Page 46 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Chapter 3
studies using PBMCs.16,29,30,48 A major variation of enzyme activity was found, ranging from 1.7 times to 500 times decreased as compared with the normal enzyme activity and once the enzyme activity was undetectable,30 although it must be mentioned that these results could be influenced by other copresent DPYD variants. Patients with DPYD*13 showed severe toxic side effects in several studies.4,24,29,44,48,49 Also dose adjustments were described by two groups.4,24 Morel et al. described a heterozygous patient that experienced severe grade 4 toxicity. After a 6-week treatment interruption, 5-FU was safely reintroduced with individual pharmacokinetic adjustment, based on 5-FU plasma levels.4 The above mentioned studies show that DPYD*13 results in an almost nonfunctional enzyme and consequently low enzyme activity levels. Without a dose reduction toxicities are likely to develop, however safe use of 5-FU is still possible with a dose adjustment. We suggest a starting dose of 50% for patients carrying DPYD*13 to ensure safe and effective use of fluoropyrimidines.
c.1236G>A/HapB3 (rs56038477)
The c.1236G>A variant was first described by Seck et al., as a silent mutation that displays normal DPD enzyme activity.46 The c.1236G>A polymorphism occurs in exon 11 and is a synonymous variant that is in complete linkage with c.483+18G>A, c.680+139G>A, c.959- 51T>G and c.1129-5923C>G;14 these variants in linkage have been termed haplotype B3.14,15 The c.1129-5923C>G intronic polymorphism (rs75017182) results in aberrant splicing and is likely to be the responsible variant for the effect on DPD enzyme activity.3,14 The frequency of heterozygous patients in Caucasian populations was reported to vary between 2.6 and 6.3%.14,15,42,49,50 DPD enzyme activity for c.1236G>A carriers was measured in PBMCs in two studies.14,46 Enzyme activities were reported to be 2.9, 4.2, 6.2 and 1.6 nmol/(mg*h) (normal value=9.6±2.6 nmol/[mg*h]) for one homozygous and three heterozygous carriers of c.1236G>A, respectively.14 In addition, a heterozygous patient in another study was found to have an enzyme activity of 10.2 nmol/(mg*h), which was reported as ‘normal activity’, since the enzyme activity of the population ranged from 4.8─15 nmol/(mg*h).46 Unfortunately data on c.1236G>A and enzyme activity are limited and not consistent. The homozygous patient still had 30% DPD activity remaining.18 Furthermore we observed two homozygous patients with this variant in our own institute with a relevant DPD enzyme activity left of around 50%, showing that this variant does not result in a completely nonfunctional enzyme (author’s unpublished data). In the study of Sistonen et al. the ratio between endogenous dihydrouracil (DHU) and uracil (U) was measured in patients carrying the c.1129-5923C>G variant.50 This ratio can be used as a phenotyping marker for DPD enzyme activity, as described in several studies.51-55 Sistonen et al. found a statistically significant decrease in DHU/U-ratio compared with wild-type patients (p=0.044). However, no significant effect for the other DPYD risk variants (DPYD*2A, DPYD*13 and c.2846A>T) was observed, which might be caused by the small sample size of patients with those variants. The c.1236G>A/ HapB3 variant has been associated with severe and lethal toxicity.14,15,42,49,56 For example, Froehlich et al. found a relative risk of 3.74 (p=2x10-5) in c.1236G>A/HapB3 carriers for severe toxicity (grade 3─5).49 In contrast, no significant effect of the c.1236G>A/HapB3 variant was found in two other studies.44,47 A dose reduction to prevent toxicity may be advantageous since multiple studies found a correlation with severe toxicity; however the
44