Gemcitabine is rapidly metabolized to its inactive metabolite, 2′,2′-difluorodeoxyuridine (dFdU), by cytidine deaminase (CDA). We previously reported that a patient with homozygous 208A alleles of CDA showed severe adverse reactions with an increase in gemcitabine plasma level. This study extended the investigation of the effects of CDA genetic polymorphisms on gemcitabine pharmacokinetics and toxicities.

Genotyping of CDA was performed by a direct sequencing of DNA obtained from the peripheral blood of Japanese gemcitabine-naïve cancer patients (n = 256). The patients recruited to the association study received a 30-minute intravenous infusion of gemcitabine at a dose of either 800 or 1,000 mg/m2, and eight blood samples were periodically collected (n = 250). Plasma levels of gemcitabine and dFdU were measured by high-performance liquid chromatography. Plasma CDA activities toward cytidine and gemcitabine were also measured (n = 121).

Twenty-six genetic variations, including 14 novel ones and two known nonsynonymous single nucleotide polymorphisms (SNPs), were detected. Haplotypes harboring the nonsynonymous SNPs 79A>C (Lys27Gln) and 208G>A (Ala70Thr) were designated *2 and *3, respectively. The allelic frequencies of the two SNPs were 0.207 and 0.037, respectively. Pharmacokinetic parameters of gemcitabine and plasma CDA activities significantly depended on the number of haplotype *3. Haplotype *3 was also associated with increased incidences of grade 3 or higher neutropenia in the patients who were coadministered fluorouracil, cisplatin, or carboplatin. Haplotype *2 showed no significant effect on gemcitabine pharmacokinetics.

Haplotype *3 harboring a nonsynonymous SNP, 208G>A (Ala70Thr), decreased clearance of gemcitabine, and increased incidences of neutropenia when patients were coadministered platinum-containing drugs or fluorouracil.

Gemcitabine (2′,2′-difluorodeoxycytidine) is a nucleoside anticancer drug that has a broad spectrum of antitumor activity against various solid tumors, such as non–small-cell lung cancer and pancreatic cancer.1 In a randomized clinical trial, gemcitabine was confirmed to provide a survival advantage over fluorouracil in addition to symptom-relieving benefits in patients with advanced pancreatic cancer.2 On the basis of these results, gemcitabine has generally been accepted as a standard chemotherapeutic agent for advanced pancreatic cancer.

Gemcitabine is transported into cells by concentrative and equilibrative nucleoside transporters,3-8 where it is phosphorylated to its monophosphate form by deoxycytidine kinase. Gemcitabine triphosphate, an active form of gemcitabine, is incorporated into an elongating DNA strand, and is followed by the addition of another deoxynucleotide that leads to the halt of DNA synthesis.9,10 Another mode of action in solid tumors, associated with the inhibition of ribonucleotide reductase, has also been suggested.11

Gemcitabine is rapidly metabolized to an inactive metabolite, 2′,2′-difluorodeoxyuridine (dFdU) by cytidine deaminase (CDA),9 and most of an administered dose is recovered as dFdU in the urine.12 CDA is expressed at varying levels in the human tissues,13 and the rapid clearance of gemcitabine can be attributed to its plentiful occurrence in the liver.14 Two single nucleotide polymorphisms (SNPs), 79A>C (Lys27Gln) and 435T>C (Thr145Thr), have been discovered in CDA, the CDA-encoding gene in humans.15,16 The 79A>C SNP reportedly reduces the deamination activity (maximum velocity/Km) toward 1-beta-D-arabinofuranosyl cytosine (cytarabine),15 and increases Km toward gemcitabine,17 in vitro. A recently discovered third SNP, 208G>A (Ala70Thr) displayed a decrease in deamination activity of 60% for cytidine and 68% for cytarabine when introduced into a CDA-null yeast strain.18

Toxicity of gemcitabine is generally mild,19,20 but unpredictable severe toxicities such as myelosuppression are occasionally experienced.21,22 Our previous case report described a patient with homozygous 208A alleles of the CDA gene who showed severe adverse reactions with increased plasma gemcitabine levels.23 In addition, there has been controversy over the relationship between cellular CDA activity and the clinical effects of cytarabine.24-27 This study examined the relationship between CDA polymorphisms, and the pharmacokinetics of gemcitabine, plasma CDA activity, or adverse reactions in Japanese cancer patients.

Gemcitabine and dFdU for analytic standards were supplied by Eli Lilly Japan K.K. (Kobe, Japan). Tetrahydrouridine, 3′-deoxy-3′-fluoro-thymidine (3′-dFT), cytidine and uridine (Sigma-Aldrich Chemical Co, St Louis, MO) were purchased. All other chemicals were of highest grade available.


The participants in this study consisted of 256 Japanese patients with carcinoma, including six patients described in a previous report,23 at the National Cancer Center Hospital (Tokyo, Japan) and National Cancer Center Hospital East (Kashiwa, Japan). Two hundred fifty-one patients received a 30-minute intravenous infusion of gemcitabine at a dose of either 800 or 1,000 mg/m2, and five patients received a fixed dose-rate (10 mg/m2/min) infusion at a dose between 1,000 and 1,500 mg/m2. The eligibility criteria for the study were as previously reported.23 The ethics committees of the National Cancer Center and the National Institutes of Health Sciences approved this study. Written informed consent was obtained from each participant.

Monitoring and Toxicities

A complete medical history and data on physical examinations were recorded before the gemcitabine therapy. CBC and platelet counts, as well as blood chemistry, were measured once a week during the first 2 months of gemcitabine treatment. Toxicities were graded according to the National Cancer Institute Common Toxicity Criteria, version 2.

DNA Sequencing

All four exons and the 5′-upstream region (approximately 800 base pairs [bp] from the translation initiation codon) of CDA were amplified from 100 ng of DNA extracted from peripheral blood, and sequenced along both strands. Polymerase chain reaction (PCR) primers23 and sequencing and PCR conditions28 were described previously. For detection of an approximately 300-bp Alu insertion (IVS3-194_-193insAlu), PCR was performed using a specific primer set (5′- TTGTCATAGCAGAAGGAGGTT-3′ and 5′- TCAGCTCTCCACACCATAAGG-3′) and 100 ng of DNA as a template. Then, sizes of the amplified fragments were determined by 1% agarose gel electrophoresis. NT_004610.17 (GenBank, National Center for Biotechnology Information, Bethesda, MD) was used as the reference sequence.

Linkage Disequilibrium and Haplotype Analyses

Hardy-Weinberg equilibrium and linkage disequilibrium (LD) analyses were performed by SNPAlyze software (Dynacom Co, Yokohama, Japan). All of the detected variations were found to be in Hardy-Weinberg equilibrium (P ≥ .05), except for the SNP IVS1+37G>A (P = .002). Some of the haplotypes were unambiguously assigned from subjects with homozygous variations at all sites or a heterozygous variation at only one site. The diplotype configurations (a combination of haplotypes) were separately inferred by LDSUPPORT software,29 which determines the posterior probability distribution of the diplotype configuration for each subject based on the estimated haplotype frequencies. The diplotype configurations of all but 11 subjects were inferred with probability of more than 0.93. All haplotypes inferred in single subjects were gathered as the groups “Other *1” and “Other *2” in Table 1.

Pharmacokinetic Study

Five patients with fixed dose-rate infusion and one patient with interruption of infusion for more than 15 minutes were excluded from the pharmacokinetic analysis described herein. Heparinized blood was collected before administration of gemcitabine and used to measure plasma CDA activity. Five milliliters of heparinized blood was also sampled for pharmacokinetic analysis before the first gemcitabine administration, and at 0, 15, 30, 60, 90, 120, and 240 minutes after the termination of the infusion. Fifty microliters of 1% tetrahydrouridine was immediately added to these samples to prevent ex vivo deamination. Plasma levels of gemcitabine and dFdU were determined using the high-performance liquid chromatography method previously reported.23 The area under the curve (AUC) and mean residence time from 0 to infinity, peak concentration (Cmax), clearance (CL/m2) and distribution volume based on the terminal phase (Vz/m2) were calculated using WINNonlin (Scientific Consultant, Apex, NC) version 4.01 (Pharsight Corporation, Mountain View, CA). AUC and Cmax were corrected for dose, assuming that all patients received 1,000 mg/m2 of gemcitabine.

CDA Activities in Plasma

Determination of CDA activities was performed using the method by Richards et al30 with slight modifications (modifications are as follows: gemcitabine was used as a substrate as well as cytidine, internal standards for analysis [3′-dFT for gemcitabine or dFdU for cytidine] were added to the mixtures at the beginning of the reaction, and high-performance liquid chromatography was used for detection of reaction products). CDA activity was expressed by unit, and one unit of enzyme activity was defined as the concentration that produces 0.1 nmol of dFdU or uridine per minute per milliliter of plasma.30

Statistical Analysis

Kruskal-Wallis, Mann-Whitney, and Pearson's correlation tests were performed using the JMP software (SAS Institute Inc, Cary, NC). Two ordinally scaled categoric data were subjected to χ2 analysis for a correlation test. A significance level of .05 was applied to all two-tailed and correlation tests. Multiplicity was adjusted by the false-discovery rate,31 if necessary.

Genetic Variations and Haplotype Structures of CDA

Twenty-six (14 novel) genetic variations were detected in the 256 Japanese cancer patients enrolled onto this study (Table 2). Three of the novel variations were found in the 5′-untranslated region, one in exon 2, three in the 3′-untranslated region and seven in the introns. Three known SNPs in the coding region of CDA were also detected. Among these, the nonsynonymous SNPs, 79A>C (Lys27Gln) and 208G>A (Ala70Thr), exhibited allelic frequencies of 0.207 and 0.037 (Table 2), respectively, and they were comparable to those reported previously.18 One patient was found to be homozygous for the 208A polymorphism. A novel insertion of an approximately 320-bp Alu element (IVS3-194_-193insAlu) was newly found in intron 3.

The detected variations were used to analyze LD (Fig 1). Four novel variations (IVS3-56G>A, IVS3-36G>A, IVS3-23C>T and *196_*197insC), the Alu element insertion and a known SNP 435C>T (Thr145Thr) showed complete linkage (Fig 1) with a frequency of 0.293. Strong LD (r2 ≥ 0.93) was also observed among SNPs −451C>T, −92A>G, and 79A>C. Note that moderate linkages (r2 ≥ 0.42) were observed between the two completely and strongly linked groups (Fig 1). Because relatively close linkages were observed throughout the entire CDA gene spanning approximately 30 kb, the CDA haplotypes were analyzed as one LD block.

The haplotypes determined/inferred in this study are summarized in Table 1. Haplotypes without amino acid changes were defined as the *1 group. These harboring the nonsynonymous SNPs 79A>C and 208G>A were designated *2 and *3, respectively. The most frequent haplotype was *1a (frequency, 0.342), followed by *2a (0.164), *1b (0.123), and *1c (0.102).

Effects of Patient Background Factors on Gemcitabine Pharmacokinetics

Characteristics of the 250 patients recruited for the pharmacokinetic study are shown in Table 3. As previously reported, the patient who was homozygous for 208A showed extraordinarily high gemcitabine and low dFdU plasma concentrations.23 Therefore, this patient was excluded when effects of patient background factors on the pharmacokinetic parameters of gemcitabine were analyzed.

The effects of age and sex on pharmacokinetic parameters are summarized in Table 4. Vz/m2 was significantly higher in males than in females, even after adjustments for their body surface areas (Mann-Whitney P = .0031). The Cmax, AUC, CL/m2, and Vz/m2 of gemcitabine showed significant correlations with age (P < .0001 for all parameters). Values of any clinical tests, including creatinine concentration, did not correlate with pharmacokinetic parameters of gemcitabine. Although approximately 30% of patients in this study underwent combined chemotherapy, no clinically significant effects of coadministered drugs on pharmacokinetic parameter values of gemcitabine were detected.

Effects of CDA Genetic Polymorphisms on Gemcitabine Pharmacokinetics

Because age and sex were unbiasedly distributed among the patients, with the various genotypes compared in the following analysis (data not shown), the 250 patients were not further stratified.

After careful examination, the data did not identify any *1, *2, or *3 subtypes that showed statistically significant differences from each major subtype within the three groups (Table 5; unpublished data). Therefore, each subtype was combined into one group (the *1, *2, or *3 group) to investigate the association between pharmacokinetic parameters and genetic groups.

The relationships between the diplotype groups and the pharmacokinetic parameters of gemcitabine are shown in Figure 2 and summarized in Table 6. The data clearly showed a haplotype *3–dependent decrease in clearance and increases in Cmax and AUC values (χ2 trend P < .0001 for all parameters). The values of Cmax, AUC, and CL/m2 observed in the patient bearing a homozygous 208G>A (*3/*3) were two-fold, five-fold, and one-fifth of the means of the *1/*1 group, respectively (Table 6). In contrast, the pharmacokinetic parameters of gemcitabine except for mean residence time (data not shown) were not significantly influenced by the haplotype *2.

Effect of Haplotypes *2 and *3 on Plasma CDA Activity

Plasma CDA activities were measured in 121 patients of the 250 patients in this study. One patient in the *1/*2 group who showed extremely high plasma CDA activities to both gemcitabine and cytidine (43.04 and 29.04 units, respectively; far higher than the 99% upper confidence limits of plasma CDA activities for the *1/*2 group) was excluded as an outlier from the following statistical analysis, although his pharmacokinetic parameters were quite normal.

Haplotype *2 failed to show any significant effects on the plasma CDA activities toward both gemcitabine and cytidine. On the other hand, activity decreased depending on the number of haplotype *3 (Table 6; Fig 3). The plasma CDA activities in the homozygous *3 (208A) patient were 12% (gemcitabine) and 25% (cytidine) of the median activities for the *1/*1 patients. As shown in Figure 4, a statistically significant correlation between the plasma CDA activity toward gemcitabine and the AUC values of gemcitabine was observed (r = −0.30; P = .0009). However, the correlations were not remarkable.

Effect of Haplotype *3 on Toxicities

Then, associations of haplotype *3 with toxicities were analyzed. Nadir grades of neutrophil counts were compared between the patient groups with and without haplotype *3 under the individual therapeutic regimens. As shown in Table 7, there were no significant differences in incidences of grade 3 or higher neutropenia between the two groups under the gemcitabine monotherapy. However, when gemcitabine was administered with carboplatin, cisplatin, or fluorouracil, grade 3 or higher neutropenia was more frequently observed in the haplotype *3–bearing group than in the group without haplotype *3. The increases in incidences were statistically significant. AUC values were also increased in the group with haplotype *3 under concomitant therapeutic regimen as under the monotherapy.

The pharmacokinetic parameters summarized in Table 4 showed great similarity to those obtained with adult American patients.32 The age-dependent decrease in gemcitabine clearance in Japanese patients in this study is in agreement with the description for Gemzar injections (Eli Lilly Japan K.K.), which is based on a population pharmacokinetic study performed outside Japan. The main route of gemcitabine elimination is its metabolism into dFdU, and there was no correlation between plasma creatinine level and gemcitabine clearance. Therefore, the aging effect on gemcitabine clearance is likely to result from a decrease in distribution volume or liver function. It is also indicated on the label that the elimination half-life of gemcitabine was longer in females than in males in a population pharmacokinetic study using 45 Japanese non–small-cell lung cancer patients. The present study did not reveal any significant sex-based difference in clearance. However, the distribution volume was significantly smaller in females than in males.

Human CDA is involved in the salvaging of pyrimidines,33,34 and plays a key role in detoxifying gemcitabine. Although the activities of 27Gln or 70Thr variant (the products of 79A>C or 208G>A) toward cytidine and cytarabine were reported to be lower than those of the “prototype” in a yeast expression system,18 the decreased CDA activity in patients bearing these SNPs has not been reported. Kreis et al35 reported that the response of leukemic patients to cytarabine correlated with the phenotype of CDA deamination determined based on the ratio of plasma concentrations of a cytarabine metabolite and cytarabine.35 They reported that 70% of subjects were slow metabolizers. However, the relationship between genetic polymorphisms and phenotypes remained to be clarified.

In our study, the haplotype *2 harboring 79C (27Gln) did not show clear effects on the AUC and CL/m2 values. In contrast, the 208A (Thr70, *3) -dependent decreases in gemcitabine clearance and plasma CDA activities were clearly demonstrated in this study. These results suggest that the CDA variant loses its in vivo deamination activities toward gemcitabine considerably. Moreover, the decreased plasma CDA activities toward gemcitabine and cytidine ex vivo also strongly suggest that the reduced enzymatic activity was caused by the genetic variation.

In the monotherapy group, the increased AUC in the patient with haplotype *3 did not clearly augment the incidence of toxicities including neutropenia. However, the incidences of grade 3 or higher neutropenia were higher in patients heterozygous for haplotype *3 compared with in the patients without haplotype *3 when they received concomitant chemotherapy with fluorouracil or platinum compounds. As we reported recently, one patient homozygous for haplotype *3 who received both gemcitabine and cisplatin suffered from extremely severe adverse effects including grade 3 anathema.23 However, he experienced neither of the specific toxicities associated with cisplatin, nephrotoxicity, and neurotoxicity. Abbruzzese et al36 reported the gemcitabine dose-dependent increase in incidence of thrombocytopenia (one of seven at 525 mg/m2/wk, three of nine at 790 mg/m2/wk, and three of six at 1,000 mg/m2/wk).36 Therefore, we concluded that extremely high exposure to gemcitabine (AUC five times higher than the average) due to the decreased deamination activity caused the life-threatening severe toxicities in this patient. In contrast, the gemcitabine AUC of the patients with heterozygous haplotype *3 was only slightly (23% to 48%) increased from that of the patients having no haplotype *3 (Table 6). This finding coincides with the lack of life-threatening severe toxicities in the heterozygotes for *3, although the incidences of grade 3 or higher neutropenia in the heterozygotes in combined chemotherapy groups were higher in the group without haplotype *3.

CDA is also involved in the activation of capecitabine to its active form fluorouracil.37 Therefore, capecitabine activation would be inefficient in patients who are homozygous for 208A. The allele frequency of the 208G>A SNP, a tagging SNP of haplotype *3, was reported to be 0.125 in Africans, while it was not detected in Europeans.38 The frequency of homozygous carriers of the variant could be higher in Africans than in the Japanese population. However, the frequency of 208G>A in Africans is still controversial, because it was not detected in 60 African Americans in a recent report.17 Extra attention may be necessary for patients with the allele before treatments with gemcitabine or cytarabine are initiated, especially to *3/*3 patients, although more studies are necessary to confirm the clinical importance of this allele in the treatments using gemcitabine or cytarabine.

A number of studies have investigated the associations between cellular CDA activity and drug responses to cytarabine.24-27,39 However, correlation between plasma CDA activity and the pharmacokinetics of gemcitabine has not been reported. Plasma CDA activity may be a useful biomarker to screen patients with a markedly decreased metabolic CDA activity such as the patient homozygous for the *3 allele found in our study, who showed extremely low plasma CDA activity. However, a very low contribution of plasma CDA to the total clearance of gemcitabine was reported,36 and the plasma CDA levels are increased in the inflammatory diseases.30,40 These may account for the failure in obtaining good correlations between plasma CDA activity and the pharmacokinetic parameters of gemcitabine, as shown in Figure 4.

In conclusion, we analyzed the CDA genetic variations and haplotypes in Japanese cancer patients who received gemcitabine. We then investigated the associations between genetic polymorphisms and the pharmacokinetics of gemcitabine or toxicities. Depending on the haplotype *3 harboring 208A, the metabolic clearance of gemcitabine decreased, and AUC and Cmax values were increased. Moreover, plasma CDA activities correlated well with the CDA genotypes. The clinical importance of the SNP 208G>A, especially of homozygotes, should be confirmed by prospective clinical studies because only one homozygous *3 patient was found in this study.

Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment: N/A Leadership: N/A Consultant: N/A Stock: N/A Honoraria: Nagahiro Saijo, Chugai, AstraZeneca, Bristol-Myers Squibb Co Research Funds: Nagahiro Saijo, Bristol-Myers Squibb Co Testimony: N/A Other: N/A

Conception and design: Nahoko Kaniwa, Shogo Ozawa, Jun-ichi Sawada, Naoyuki Kamatani, Hideki Ueno, Takuji Okusaka, Nagahiro Saijo

Financial support: Jun-ichi Sawada, Teruhiko Yoshida, Nagahiro Saijo

Administrative support: Nahoko Kaniwa, Ryuichi Hasegawa, Yoshiro Saito, Shogo Ozawa, Jun-ichi Sawada, Teruhiko Yoshida, Nagahiro Saijo

Provision of study materials or patients: Keiko Maekawa, Yoshiro Saito, Shogo Ozawa, Junji Furuse, Hiroshi Ishii, Hideki Ueno, Takuji Okusaka

Collection and assembly of data: Emiko Sugiyama, Su-Ryang Kim, Ruri Kikura-Hanajiri, Keiko Maekawa

Data analysis and interpretation: Emiko Sugiyama, Nahoko Kaniwa, Su-Ryang Kim, Yoshiro Saito, Junji Furuse, Hiroshi Ishii, Hideki Ueno, Takuji Okusaka

Manuscript writing: Emiko Sugiyama, Nahoko Kaniwa, Su-Ryang Kim, Hideki Ueno

Final approval of manuscript: Nahoko Kaniwa, Jun-ichi Sawada, Hideki Ueno, Nagahiro Saijo


Table 1. CDA Haplotypes Estimated in This Study


Table 1. CDA Haplotypes Estimated in This Study (continued>


Table 2. Variations of the CDA Gene Found

Table 2. Variations of the CDA Gene Found

Nucleotide Change and Flanking Sequences (5′ to 3′)Amino Acid ChangeAllele Frequency
This StudyNCBI (dbSNP)JSNPNT_004610.17From the Translational Initiation Site or From the Nearest Exon
MPJ6_CDA005rs602950Exon 1 (5′-UTR)3739873−92GGGACACACCCAA/GGGGGAGGAGCTG0.205
MPJ6_CDA007rs3215400IMS-JST076939Exon 1 (5′-UTR)3739934−33_−31GCTCCTGTTTCCC/-GCTGCTCTGCTG0.451
MPJ6_CDA009rs2072671IMS-JST008769Exon 1374004379CAGGAGGCCAAGA/CAGTCAGCCTACTLys27Gln0.207
MPJ6_CDA010rs12059454Intron 13740155IVS1+37CCCAGCCCAGCAG/ACCTGGGTGGTGG0.184
MPJ6_CDA014*Intron 23755961_3755962IVS2+87_+88TCATTCATTCAT-/TCATCTGACATATGTT0.135
MPJ6_CDA016rs10916825Intron 23756116IVS2+242CATACAAGGGCCA/GGTATGCCCCTGT0.289
MPJ6_CDA017rs818194Intron 23756170IVS2+296GTCCTACAAGATT/ATAACAGAAAGGC0.217
MPJ6_CDA018rs3738130IMS-JST083844Intron 33764805IVS3+71AGCCACGCCAAGT/CTGCAGGCATGGC0.053
MPJ6_CDA019*Intron 33769093_3769094IVS3-194_−193CTGTTCAGTTTC-/(Alu)§ACAGCATTCTTT0.293
MPJ6_CDA023rs17846527Exon 43769397435CCTGCAGAAGACC/TCAGTGACAGCCAThr145Thr0.293
MPJ6_CDA024*Exon 4 (3′-UTR)3769472510 (*69)CTCACAGCCCTGG/TGGACACCTGCCC0.002
MPJ6_CDA025*Exon 4 (3′-UTR)3769599_3769600637_638 (*196_197)ACCGCCGCCCCC-/CTGCCCCACCTTT0.293
MPJ6_CDA026*Exon 4 (3′-UTR)3769638676 (*235)GGGCCCTCTTTCA/GAAGTCCAGCCTA0.010

*Novel variations detected in this study.

†Yue et al.18

‡A of the translation initiation codon ATG is numbered 1, and the number with * in parentheses indicates the position from the termination codon TGA.



Table 3. Characteristics of Patients Recruited to Pharmacokinetic Studies (N = 250)

Table 3. Characteristics of Patients Recruited to Pharmacokinetic Studies (N = 250)

Age, years
Body surface area, m2
Weight, kg
Performance status
Primary tumor
Dose, mg/m2
    Gemcitabine alone180
    Gemcitabine-based combination70
    Vinorelbine ditartrate10
Previous treatment

Table 4. Effects of Patient Background Factors on Pharmacokinetic Parameters of Gemcitabine

Table 4. Effects of Patient Background Factors on Pharmacokinetic Parameters of Gemcitabine

FactorCmax (μg/mL)
AUC (hr · μg/mL)
CL/m2 (L/hr/m2)
Vz/m2 (L/m2)
Median1/4-3/4 QuantilesMedian1/4-3/4 QuantilesMedian1/4-3/4 QuantilesMedian1/4-3/4 Quantiles
    Mann-Whitney U testNSNSNSP < .005
    Spearman r0.320.39−0.39−0.39
    P value< .0001< .0001< .0001< .0001

Abbreviations: Cmax, peak concentration; AUC, area under the curve; CL/m2, clearance; Vz/m2, distribution volume based on the terminal phase.

*Significantly different from the value for female (Mann-Whitney U test P = .0031).


Table 5. Pharmacokinetic Parameters of Gemcitabine in Patients With Various CDA Diplotypes

Table 5. Pharmacokinetic Parameters of Gemcitabine in Patients With Various CDA Diplotypes

DiplotypeNo. of PatientsMedian Gemcitabine PK Parameters
Cmax (μg/mL)AUC (hr · μg/mL)CL/m2 (L/hr/m2)MRT (hours)AUC Ratio (dFdU/gemcitabine)
P value*0.820.400.590.970.83
P value*0.770.570.940.970.83
P value*0.820.450.900.860.57
P value0.660.980.760.0770.46

Abbreviations: PK, pharmacokinetics; Cmax, peak concentration; AUC, area under the curve; CL/m2, clearance; MRT, mean residence time; dFdU, 2′,2′-difluorodeoxyuridine.

*P value of a correlation test among *1a/*1a, *1a/(*1b, *1c, or *1d), and (*1b, *1c, or *1d)/(*1b, *1c, or *1d). Multiplicity is adjusted by false-discovery rate.

P value of a Kruskal-Wallis test among *2a/*2a, *2a/*2b, and *2a/*2c.


Table 6. Pharmacokinetic Parameters of Gemcitabine and Plasma CDA Activities in the Patient Groups Categorized According to Diplotypes

Table 6. Pharmacokinetic Parameters of Gemcitabine and Plasma CDA Activities in the Patient Groups Categorized According to Diplotypes

DiplotypeMedian Gemcitabine PK Parameters
Median CDA Activity (units)
No. of PatientsCmax (μg/mL)AUC (hr·μg/mL)CL/m2 (L/hr/m2)No. of PatientsGemcitabineCytidine
P value*0.520.460.990.470.19
P value5.94E-046.66E-137.77E-049.35E-052.45E-04

Abbreviations: CDA, cytidine deaminase; Cmax, peak concentration; AUC, area under the curve; CL/m2, clearance.

*P value of a correlation test among *1/*1, *1/*2, and *2/*2. Multiplicity is adjusted by false-discovery rate.

P value of a correlation test among *1/*1, *1/*3, and *3/*3. Multiplicity is adjusted by false-discovery rate.


Table 7. Comparison of Adverse Reaction Incidence and Pharmacokinetic Parameters of Gemcitabine Between Two Patient Groups With and Without Haplotype *3

Table 7. Comparison of Adverse Reaction Incidence and Pharmacokinetic Parameters of Gemcitabine Between Two Patient Groups With and Without Haplotype *3

ChemotherapyGenotypeIncidence of Neturopenia (nadir)*
AUC (hr·μg/mL)
≥ Grade 3
≥ Grade 4
No. of CasesTotal No. of PatientsProbabilityNo. of CasesTotal No. of PatientsProbability
Monotherapynon *3/non *3661670.408670.059.91
non *3/*36100.601100.1013.13
With fluorouracilnon *3/non *33120.252120.178.11
non *3/*3221.00120.5011.98
With carboplatinnon *3/non *39130.691130.089.87
non *3/*3331.00230.6712.48
With cisplatinnon *3/non *38280.292280.079.53
non *3/*3111.00010.0011.71

Note. No analyses were performed in patients who received gemcitabine with vinorelbine, because only one patient bore the haplotype *3. Boldfacing indicates a statistically significant difference (P < .05).


†Kruskal-Wallis test.

‡A P value for comparison between non*3/non*3 and (non*3/*3 + *3/*3).

© 2006 by American Society of Clinical Oncology

Supported by the Program for the Promotion of Fundamental Studies in Health Sciences (Grant No. MPJ6 and 05-25), and the Health and Labour Sciences Research Grant on Human Genome and Tissue Engineering (Grant No. H16-Genome-008) from the Ministry of Health, Labour, and Welfare of Japan.

Presented in part at the 41st Annual Meeting of the American Society of Clinical Oncology, May 13-17, 2005, Orlando, FL, and at the 13th Annual Meeting of the North American Society for the Study of Xenobiotics, October 22-27, 2005, Maui, HI.

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

We thank Emiko Jimbo, Miho Akimoto, Atsuko Watanabe, Tomoko Chujo, Makiyo Iwamoto, and Mamiko Shimada for assistance in sample collection and management, and Chie Sudo for secretarial assistance.

1. Noble S, Goa KL: Gemcitabine: A review of its pharmacology and clinical potential in non-small cell lung cancer and pancreatic cancer. Drugs 54::447,1997-472, Crossref, MedlineGoogle Scholar
2. Burris HA III, Moore MJ, Andersen J, et al: Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: A randomized trial. J Clin Oncol 15::2403,1997-2413, LinkGoogle Scholar
3. Rauchwerger DR, Firby PS, Hedley DW, et al: Equilibrative-sensitive nucleoside transporter and its role in gemcitabine sensitivity. Cancer Res 60::6075,2000-6079, MedlineGoogle Scholar
4. Mackey JR, Mani RS, Selner M, et al: Functional nucleoside transporters are required for gemcitabine influx and manifestation of toxicity in cancer cell lines. Cancer Res 58::4349,1998-4357, MedlineGoogle Scholar
5. Mackey JR, Yao SY, Smith KM, et al: Gemcitabine transport in xenopus oocytes expressing recombinant plasma membrane mammalian nucleoside transporters. J Natl Cancer Inst 91::1876,1999-1881, Crossref, MedlineGoogle Scholar
6. Baldwin SA, Yao SY, Hyde RJ, et al: Functional characterization of novel human and mouse equilibrative nucleoside transporters (hENT3 and mENT3) located in intracellular membranes. J Biol Chem 280::15880,2005-15887, Crossref, MedlineGoogle Scholar
7. Mangravite LM, Badagnani I, Giacomini KM: Nucleoside transporters in the disposition and targeting of nucleoside analogs in the kidney. Eur J Pharmacol 479::269,2003-281, Crossref, MedlineGoogle Scholar
8. Ritzel MW, Ng AM, Yao SY, et al: Molecular identification and characterization of novel human and mouse concentrative Na+-nucleoside cotransporter proteins (hCNT3 and mCNT3) broadly selective for purine and pyrimidine nucleosides (system cib). J Biol Chem 276::2914,2001-2927, Crossref, MedlineGoogle Scholar
9. Plunkett W, Huang P, Gandhi V: Preclinical characteristics of gemcitabine. Anticancer Drugs 6::S7,1995-S13, (suppl 6) Google Scholar
10. Plunkett W, Huang P, Searcy CE, et al: Gemcitabine: Preclinical pharmacology and mechanisms of action. Semin Oncol 23::S3,1996-S15, (suppl 10) MedlineGoogle Scholar
11. Heinemann V, Xu YZ, Chubb S, et al: Inhibition of ribonucleotide reduction in CCRF-CEM cells by 2′,2′-difluorodeoxycytidine. Mol Pharmacol 38::567,1990-572, MedlineGoogle Scholar
12. Kiani A, Kohne CH, Franz T, et al: Pharmacokinetics of gemcitabine in a patient with end-stage renal disease: Effective clearance of its main metabolite by standard hemodialysis treatment. Cancer Chemother Pharmacol 51::266,2003-270, MedlineGoogle Scholar
13. Watanabe S, Uchida T: Expression of cytidine deaminase in human solid tumors and its regulation by 1 alpha, 2, 5-dihydroxyvitamin D3. Biochim Biophys Acta 1312::99,1996-104, Crossref, MedlineGoogle Scholar
14. Ho DH: Distribution of kinase and deaminase of 1-beta-D-arabinofuranosylcytosine in tissues of man and mouse. Cancer Res 33::2816,1973-2820, MedlineGoogle Scholar
15. Kirch HC, Schroder J, Hoppe H, et al: Recombinant gene products of two natural variants of the human cytidine deaminase gene confer different deamination rates of cytarabine in vitro. Exp Hematol 26::421,1998-425, MedlineGoogle Scholar
16. Schroder JK, Kirch C, Seeber S, et al: Structural and functional analysis of the cytidine deaminase gene in patients with acute myeloid leukaemia. Br J Haematol 103::1096,1998-1103, Crossref, MedlineGoogle Scholar
17. Gilbert JA, Salavaggione OE, Ji Y, et al: Gemcitabine pharmacogenomics: Cytidine deaminase and deoxycytidylate deaminase gene resequencing and functional genomics. Clin Cancer Res 12::1794,2006-1803, Crossref, MedlineGoogle Scholar
18. Yue L, Saikawa Y, Ota K, et al: A functional single-nucleotide polymorphism in the human cytidine deaminase gene contributing to ara-C sensitivity. Pharmacogenetics 13::29,2003-38, Crossref, MedlineGoogle Scholar
19. Aapro MS, Martin C, Hatty S: Gemcitabine: A safety review. Anticancer Drugs 9::191,1998-201, Crossref, MedlineGoogle Scholar
20. Gallelli L, Nardi M, Prantera T, et al: Retrospective analysis of adverse drug reactions induced by gemcitabine treatment in patients with non-small cell lung cancer. Pharmacol Res 49::259,2004-263, Crossref, MedlineGoogle Scholar
21. Bokemeyer C, Gerl A, Schoffski P, et al: Gemcitabine in patients with relapsed or cisplatin-refractory testicular cancer. J Clin Oncol 17::512,1999-516, LinkGoogle Scholar
22. Locker GJ, Wenzel C, Schmidinger M, et al: Unexpected severe myelotoxicity of gemcitabine in pretreated breast cancer patients. Anticancer Drugs 12::209,2001-212, Crossref, MedlineGoogle Scholar
23. Yonemori K, Ueno H, Okusaka T, et al: Severe drug toxicity associated with a single-nucleotide polymorphism of the cytidine deaminase gene in a Japanese cancer patient treated with gemcitabine plus cisplatin. Clin Cancer Res 11::2620,2005-2624, Crossref, MedlineGoogle Scholar
24. Colly LP, Peters WG, Richel D, et al: Deoxycytidine kinase and deoxycytidine deaminase values correspond closely to clinical response to cytosine arabinoside remission induction therapy in patients with acute myelogenous leukemia. Semin Oncol 14::S257,1987-S261, (suppl 1) MedlineGoogle Scholar
25. Steuart CD, Burke PJ: Cytidine deaminase and the development of resistance to arabinosyl cytosine. Nat New Biol 233::109,1971-110, MedlineGoogle Scholar
26. Tattersall MH, Ganeshaguru K, Hoffbrand AV: Mechanisms of resistance of human acute leukaemia cells to cytosine arabinoside. Br J Haematol 27::39,1974-46, Crossref, MedlineGoogle Scholar
27. Chiba P, Tihan T, Szekeres T, et al: Concordant changes of pyrimidine metabolism in blasts of two cases of acute myeloid leukemia after repeated treatment with ara-C in vivo. Leukemia 4::761,1990-765, MedlineGoogle Scholar
28. Nakamura T, Saito Y, Murayama N, et al: Apparent low frequency of sequence variability within the proximal promoter region of the cytochrome P450(CYP)3A5 gene in established cell lines from Japanese individuals. Biol Pharm Bull 24::954,2001-957, Crossref, MedlineGoogle Scholar
29. Kitamura Y, Moriguchi M, Kaneko H, et al: Determination of probability distribution of diplotype configuration (diplotype distribution) for each subject from genotypic data using the EM algorithm. Ann Hum Genet 66::183,2002-193, Crossref, MedlineGoogle Scholar
30. Richards DA, Sherwood RA, Ndebele D, et al: Determination of plasma cytidine deaminase activity by HPLC. Biomed Chromatogr 2::148,1987-151, Crossref, MedlineGoogle Scholar
31. Benjamini Y, Hochberg Y: Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Statist Soc Ser B 57::289,1995-300, Google Scholar
32. Bhargava P, Marshall JL, Fried K, et al: Phase I and pharmacokinetic study of two sequences of gemcitabine and docetaxel administered weekly to patients with advanced cancer. Cancer Chemother Pharmacol 48::95,2001-103, Crossref, MedlineGoogle Scholar
33. Johansson E, Mejlhede N, Neuhard J, et al: Crystal structure of the tetrameric cytidine deaminase from Bacillus subtilis at 2.0: A resolution. Biochemistry 41::2563,2002-2570, Crossref, MedlineGoogle Scholar
34. Costanzi S, Vincenzetti S, Vita A, et al: Human cytidine deaminase: Understanding the catalytic mechanism. Nucleosides Nucleotides Nucleic Acids 22::1539,2003-1543, Crossref, MedlineGoogle Scholar
35. Kreis W, Lesser M, Budman DR, et al: Phenotypic analysis of 1-B-D-arabinofuranosylcytosine deamination in patients treated with high doses and correlation with response. Cancer Chemother Pharmacol 30::126,1992-130, Crossref, MedlineGoogle Scholar
36. Abbruzzese JL, Grunewald R, Weeks EA, et al: A phase I clinical, plasma, and cellular pharmacology study of gemcitabine. J Clin Oncol 9::491,1991-498, LinkGoogle Scholar
37. Nishida M: Pharmacological and clinical properties of Xeloda (capecitabine), a new oral active derivative of fluoropyrimidine [Japanese]. Nippon Yakurigaku Zasshi 122::549,2003-553, Crossref, MedlineGoogle Scholar
38. Fukunaga AK, Marsh S, Murry DJ, et al: Identification and analysis of single nucleotide polymorphisms in the gemcitabine pharmacologic pathway. Pharmacogenomics J 4::307,2004-314, Crossref, MedlineGoogle Scholar
39. Jahns-Streubel G, Reuter C, Auf der Landwehr U, et al: Activity of thymidine kinase and of polymerase alpha as well as activity and gene expression of deoxycytidine deaminase in leukemic blasts are correlated with clinical response in the setting of granulocyte-macrophage colony-stimulating factor-based priming before and during TAD-9 induction therapy in acute myeloid leukemia. Blood 90::1968,1997-1976, Crossref, MedlineGoogle Scholar
40. Sherwood RA: The measurement of nucleoside deaminases by high performance liquid chromatography and their use in clinical chemistry. Biomed Chromatogr 5::235,1991-239, Crossref, MedlineGoogle Scholar
Downloaded 1,452 times


No companion articles


DOI: 10.1200/JCO.2006.06.7405 Journal of Clinical Oncology 25, no. 1 (January 01, 2007) 32-42.

Published online September 21, 2016.

ASCO Career Center