The pharmacogenetics of paclitaxel and gemcitabine

Paclitaxel binds to beta tubulin, causing stabilization of polymerized microtubules resulting in cell cycle arrest and the inhibition of mitosis. Recently, low-dose paclitaxel has demonstrated anti-angiogenic activity. A phase I trial was designed to assess the safety of escalating paclitaxel in combination with alpha interferon. Pharmacokinetic parameters of paclitaxel alone and in combination with interferon are determined using ADAPT software. Paclitaxel exhibits variable pharmacokinetics and toxicity. Polymorphisms in the paclitaxel pharmacologic pathway may help explain this interindividual variability. Seventeen subjects in this phase I trial were genotyped using PCR and pyrosequencing techniques. Nine polymorphisms in ABCB1, CYP2C8 and NR1I2 were genotyped to determine if genotype influences Cmax, AUC and clearance. Paclitaxel pharmacokinetic parameters are not affected by co-administration with interferon. Subjects who are homozygous for the variant allele, T, for ABCB1 3435 demonstrates lower paclitaxel clearance than subjects homozygous for the wild type, C, allele. Gemcitabine is a nucleoside analog that is primarily used in the treatment of pancreatic cancer and is also approved for use in NSCLC and breast cancer. Gemcitabine competes with endogenous dCTP, for incorporation during DNA synthesis. Gemcitabine demonstrates interindividual variability. Because gemcitabine has a complex pharmacologic pathway, it is not likely that a single polymorphism in a single gene can explain the variability in gemcitabine metabolism and toxicity. Polymorphisms throughout the pathway were identified and genotyped in three populations, European, African and Asian. Haplotypes of DCK and CDA, were constructed in the European, Asian and African population. Polymorphisms in the 5′FR of DCK and CDA were analyzed in 52 paired (tumor and normal) colorectal cancer samples to determine if they were significant predictors of mRNA expression. dFdC pathway SNP analysis revealed significant differences in variant allele frequencies among the three populations. Three haplotypes for DCK captured ∼90% of the genomic structure in the three populations evaluated consequently, DCK haplotypes may be a useful tool in association studies. Because of the lack of linkage disequilibrium in CDA, individual polymorphisms may offer more information on the genotype-phenotype relationship. Lastly, in the 52 colorectal cancer samples, DCK and CDA 5 ′FR genotype does not affect mRNA expression