Different patients respond in different ways to the same medication. Although many nongenetic factors, including age, concomitant therapy, drug interactions, and the nature of the disease, may influence the outcomes of medications, there are many inter-individual differences in drug response owing to the sequence variants of genes that encode drug-metabolizing enzymes, drug transporters, or drug targets [1]. Most of the drug-metabolizing enzymes and transporters have a broad range of genetic polymorphisms, which may cause inter-individual variability with different concentrations of drugs. In addition, anticancer therapies are known to have a narrow therapeutic range; a high concentration in a patient’s body increases the toxicity, and a low concentration decreases the effect of the drug [2]. Some of the single-nucleotide polymorphisms (SNPs) have already been correlated with substantial changes in the metabolism or efficacy of the drug, and some are being utilized to predict clinical outcomes [1].

Docetaxel has an effective antitumor activity against many cancers, such as locally advanced or metabolic breast cancer, non-small-cell lung cancer (NSCLC), and androgen-independent prostate cancer, and can induce response rates between 20% and 40% [3]. However, despite its widespread use, an inter-individual variability in the toxicities of docetaxel has been a major challenge in clinical practice. Dose-limiting toxicities of docetaxel are considered to be neutropenia, anemia, nausea, asthenia, and skin toxicity [4]. Among the severe toxicities, neutropenia is one of the dose-limiting adverse reactions and is often observed at a frequency of approximately 36% [4,5]. The other side effects include alopecia, asthenia, dermatologic reactions, fluid retention, hypersensitivity reactions, and stomatitis. One of the most important factors that limit the use of docetaxel use is its unpredictability of inter-individual variation in toxicity. The potential causes of variability include pathogenesis and disease severity, occurrence of unintended drug interactions, and impairment of hepatic and/or renal functions [4].

The elimination of docetaxel occurs mainly through a metabolic conversion by CYP3A4 and CYP3A5, which results in a formation of metabolites with reduced cytotoxic activity [5,6]. The biotransformation of docetaxel at the hepatic level leads to the inactivation of the molecule and reduction of its therapeutic effect [6]. Biotransformation is the main route for elimination of docetaxel, which makes it an interesting drug for the investigation of the genetic polymorphism in CYP450 enzymes [5]. Moreover, the correlation between high CYP3A4 mRNA expression and low chemical susceptibility of docetaxel in breast cancer tissues has been shown by real-time polymerase chain reaction and immunohistochemistry. A high CYP3A4 mRNA expression in tumor tissues can be predicted to accelerate the speed of docetaxel metabolism, and thus, results in resistance [7].

The elimination pathway is widely known to be mediated by the drug efflux ABC transporter, ABCB1 (also known as P-glycoprotein or MDR1) [8]. Moreover, the expression of these genes is regulated by the nuclear hormone receptors pregnane X receptor and constitutive androstane receptor encoded by NR1I2 and NR1I3, respectively [9]. Other transporters, ABCC2 (MRP2) and SLCO1B3 (OTAP1B3 or OATP8), are responsible for the efflux and uptake of docetaxel in vitro. Consequently, these genes are considered to be candidates that may affect the toxicity of docetaxel. There have been a few reports concerning the association between the toxicity of docetaxel and genetic variants of CYP3A4, CYP3A5, and ABCB1, and there were no reports on NR1I2, NR1I3, ABCC2, and SLCO1B3 [4]. In this study, we investigated the role of CYP3A4, CYP3A5, ABCB1, SLCO1B3, and ABCC2 polymorphisms on docetaxel toxicity.

Although several studies have shown an association between the polymorphisms and toxicity of anticancer drugs, such as docetaxel, the genetic determinants of the adverse reaction of docetaxel has not been elucidated. In this study, we genotyped 10 polymorphisms located in 5 candidate genes involved in the metabolism (CYP3A4 and CYP3A5) and transport (ABCB1, ABCC2, and SLCO1B3) of docetaxel to identify the genes related to the docetaxel-induced toxicity. We determined that 2677G/T in ABCB1 and rs12762549 in ABCC2 are significantly associated with severe docetaxel-induced leucopenia. In the subgroup analysis of NSCLC patients, significant associations of tumor response and G2677T/A in ABCB1 and rs11045585 in SLCO1B3 were also observed, suggesting that ABCB1 and SLCO1B3 may be largely involved in the transport of docetaxel.

Most of the genes that encode the enzymes involved in the activation and detoxification pathways are considered to be highly polymorphic [2]. In cancer treatment, drug therapy is an essential tool in disease regulation. However, the therapeutic window of anticancer drugs is generally narrow. Inter-individual variations in a drug-metabolizing capacity and drug response can complicate the outcome of the same treatment even in patients who have been diagnosed with the same disease [5].

In the present study, we found no association between the genotypes of the CYP3A4 and CYP3A5 genes and an increase in docetaxel toxicity. The disposition of docetaxel is mediated by CYP3A4 and CYP3A5 enzymes and transporters, such as ABCB1, ABCC2, and SLCO1B3 [10]. The estimated quantitative contribution of CYP3A4/5 to the metabolism of docetaxel in the human liver is 64%-93%, making it the major CYP450 subfamily in docetaxel elimination [5,11]. Previously, several studies on patients with the CYP3A4^*1A/1B (rs2740574) and CYP3A5^*1/^*3 (rs776746) haplotypes showed an association with an increased clearance and/or decreased exposure to docetaxel [10,12]. Both rs2710574 and CYP3A5^*1 alleles have shown to be in a linkage disequilibrium in Caucasians and African Americans [12]. Previous studies have shown that CYP3A4-CYP3A5 haplotypes may be correlated to certain diseases, impaired drug clearance, or undesirable adverse drug events [12]. In Asian cancer patients, some studies suggested a correlation between the inactive CYP3A5^*3 allele and docetaxel clearance. However, another study observed no association between the clearance and CYP3A5^*3 genotype status in CYP3A5^*1/^*1, CYP3A5^*1/^*3, and CYP3A5^*3/^*3 patients [13].

Generally, variants of the coding regions in CYP3A4 have 5% allele frequencies as heterozygous with the wild-type allele. Due to low allele frequencies, these coding variants are not likely to be the major cause of inter-individual differences in CYP3A-dependent clearance and limited alterations in enzyme expression or catalytic function. The most common variant, CYP3A4^*1B, an A-392G transition in the 5′-flanking region, has an allele frequency that ranges from 0% (Chinese and Japanese) to 45% (African Americans) [14]. In our population, we observed that CYP3A4^*1B and CYP3A4^*3 have 0% of allele frequency. In contrast, a linkage disequilibrium between CYP3A4^*1B and another CYP3A allele (CYP3A5^*1) may be the cause of clinical phenotype. CYP3A5 is polymorphically expressed in adults: approximately 10%-20% in Caucasians, 33% in Japanese, and 55% in African-Americans [15]. However, it is unlikely that CYP3A4 and CYP3A5 polymorphisms will be of clinical benefit in influencing docetaxel disposition in our patients.

Drug transporters consist of uptake and efflux transporters, indicating intracellular or extracellular transport directions. These transporters play an essential role as the defense mechanisms against penetration of xenobiotics or transmembrane transportation of various endogenous compounds [16]. In our study, 2677G/T in ABCB1 and rs12762549 in ABCC2 are significantly associated with severe docetaxel-induced leucopenia. Among the ABC proteins, some members, such as ABCB1, ABCC1, ABCC2, and ABCG2, are considered to contribute to multidrug resistance of cancer chemotherapy [17]. Recently, genetic polymorphisms in the ABCB1 gene and their association with the P-gp level have been investigated [18,19]. Some studies examined 15 SNPs, including 6 in the coding region of healthy Caucasians and found a significant association of polymorphism in exon 26 (C3435T) of ABCB1 with the expression levels and function of ABCB1 [18]. In a Japanese population, 8 SNPs have been identified in the region of ABCB1, and among these SNPs, –2410T>C, –1910T>C, and 692T>C are in perfect linkage disequilibrium [17]. In addition, several studies have shown that the inter-individual variations in activity and expression levels of ABCB1 and ABCC2 may have an effect on drug response and response to toxic agents. Therefore, our results are similar to those obtained in previous studies.

The SLCO1B3 gene is highly polymorphic, exhibiting high variations across the ethnic groups [20]. The allelic frequencies of the non-synonymous variants 334T>G (Ser112Ala; rs4149117) and 699G>A (Met233lle; rs7311358) display a great degree of heterogeneity across diverse ethnic populations, ranging from 41% in African-Americans to approximately 71%-90% in Caucasians and Chinese. However, we estimate that the allelic frequency of SLCO1B3 (G2677T/A; rs11045585) is almost 1% in this population compared to that of the previous studies. This study shows a significant association of tumor response with G2677T/A and SLCO1B3 in a subgroup of lung cancer patients. This study highlights the importance of SLCO1B3 polymorphic variations in influencing docetaxel disposition in cancer patients.

An important limitation associated with docetaxel use is the unpredictability of inter-individual efficacy and toxicity. In spite of the potential importance of the clinical variables in determining the drug effects, it is considered that inherited differences in metabolism and excretion can have a greater effect on the efficacy and toxicity of drugs [21]. Several studies have attempted to identify the genetic polymorphisms in genes encoding drug metabolizing enzymes and transporters accounting for the remarkable inter-individual variation. The studies are conflicting and inconclusive. For example, some researchers reported an increase in docetaxel clearance in carriers of CYP3A4 polymorphisms (rs2740574) [6,13], a finding that was not supported by others [21]. Differences in the sample size, various tumor types, different treatment regimens (single agent vs. combination of drugs interacting with docetaxel) and various ethnic groups might be an explanation for these conflicting findings. Additionally, most of these results were obtained from only a few SNPs in genes involved in docetaxel disposition [22].

Recently, Maemondo et al. [23] suggested that carboplatin and paclitaxel (CP) regimen achieved a longer progression-free survival with less toxicity, excluding moderate anemia and thrombocytopenia than single-agent docetaxel. Considering the results of this phase II and IFCT-0501 trials, they selected the CP regimen as a candidate for a future phase III trial in elderly patients with advanced NSCLC [23]. Although monotherapy with the third-generation agents has been regarded as the preferred treatment option for elderly patients with NSCLC, Quoix et al. [24] recently reported the results of IFCT-0501, a phase III study comparing a similar CP regimen with monotherapy with either vinorelbine or gemcitabine in an elderly population. They demonstrated a significant superiority of the CP regimen in terms of the efficacy (overall survival); however, severe toxicity in the CP arm, including a treatment-related death rate of 4.4%, was observed [24]. In this study, we found an association between toxicity and chemotherapy regimen as the previous studies in a multivariate logistic regression analysis, and docetaxel monotherapy might be more toxic than the combination therapy.

Our data suggest that ABCB1 (2677G/T) and SLCO1B3 (rs11055585) may be major genetic predictors of docetaxel-related adverse events in patients receiving docetaxel chemotherapy. Further studies are warranted to determine the usefulness of genotyping for docetaxel treatment.