Response to “Pharmacogenetics of Voriconazole: CYP2C19 but Also CYP3A4 Need to Be Genotyped” - The Role ofCYP3A4andCYP3A5Polymorphisms in Clinical Pharmacokinetics of Voriconazole
Voriconazole is metabolized through a complex series of pathways involving CYP2C19, CYP2C9, CYP3A4, and CYP3A5.2 That CYP2C19 polymorphisms may significantly alter the clinical pharmacokinetics of voriconazole is relatively well accepted.3,4 By comparison, the data supporting the pharmacogenetic impact of CYP3A4 polymorphisms on the clinical pharmacokinetics of voriconazole are limited.
Gautier-Veyret et al. adapt a “genetic score” in analyzing the effect of CYP3A4 and CYP2C19 polymorphisms on serum concentrations of voriconazole in 29 adult allogeneic hematopoietic stem cell transplant recipients from Grenoble, France.5 Unfortunately, this score assumes an equal level of human metabolic activity of CYP3A4 and CYP2C19 based on similar Km values directly derived from an in vitro Escherichia coli gene expression system.6 This is not likely a correct assumption, as differences between an Escherichia coli in vitro expression system and humans may exist in degrees of gene expression, posttranslational modification, epigenetic effects, and tissue microenvironmental modulation as potentially important determinants of human hepatic enzyme activity. Differential expression of hepatic drug-metabolizing enzymes may also be affected by nonhepatic factors, including inflammation.7 Giving equal weight in the genetic score to both enzymes may therefore not be valid. As the genetic score may be driven by CYP2C19 polymorphisms, the data do not convincingly demonstrate the independent role of CYP3A4 polymorphisms in determination of significantly altered human pharmacokinetics of voriconazole.
The study by He et al. examines the effect of 16 polymorphisms of CYP3A4 and CYP3A5 on the serum concentrations of voriconazole in 158 Chinese patients from Xi'an Jiaotong University with a proven or probable invasive fungal infection and who had received intravenous or oral voriconazole.8 Three categories of ranges of voriconazole serum levels were assigned: lower, normal, and higher. Only the rs4646437 T allele of CYP3A4 was found to have attained a statistically significant effect on the distributions among the three categories (P = 0.033); however, the 95% confidence intervals (CIs) around this difference were large (1.086–7.384). In attempting to account for these wide CIs one finds that the frequency of this allele is relatively low (=20%) compared to other distributions that commonly exceeded 70%. Moreover, the differences in allelic frequency distributions of other CYP3A4 and CYP3A5 polymorphisms, such as rs776746 and rs4646440 (T allele) that were not considered to significantly alter voriconazole serum concentrations, were relatively similar to those of the rs4646437 T allele of CYP3A4.
Certainly further studies are warranted that may more clearly elucidate the role of CYP3A4 and CYP3A5 polymorphisms in the clinical pharmacokinetics of voriconazole. At the present time, however, a conservative analysis of the available data leads one to conclude that there is not sufficient evidence to support recommending altered dosing of voriconazole based on CYP3A4 and CYP3A5 polymorphisms.