2238 Biol. Pharm. Bull. 30(12) 2238—2243 (2007) Vol. 30, No. 12

Population and Proton Pump Inhibitory Effects of Intravenous Lansoprazole in Healthy Japanese Males

a,b b b c d Yuichi SAKURAI, Masashi HIRAYAMA, Muneaki HASHIMOTO, Takanori TANAKA, Setsuo HASEGAWA, e f a a ,a Shin IRIE, Kiyoshi ASHIDA, Yuichiro KAYANO, Masato TAGUCHI, and Yukiya HASHIMOTO* a Graduate School of Pharmaceutical Sciences, University of Toyama; 2630 Sugitani, Toyama 930–0194, Japan: b Takeda Pharmaceutical Company Limited; 4–1–1 Dosyo-machi, Chuo-ku, Osaka 540–8645, Japan: c Ohsaki Clinic; 5–7–13 Kitashinagawa, Shinagawa-ku, Tokyo 141–0001, Japan: d Sekino Clinical Pharmacology Clinic; 3–28–3 Ikebukuro, Toshima-ku, Tokyo 171–0014, Japan: e Kyushu Clinical Pharmacology Research Clinic; 2–13–16 Jigyo, Chuo-ku, Fukuoka 810–0064, Japan: and f Saiseikai Nakatsu Hospital; 2–10–39 Shibata, Kita-ku, Osaka 530–0012, Japan. Received June 20, 2007; accepted September 18, 2007

A total of 56 healthy Japanese males were enrolled in single- or multiple- dose pharmacokinetic trials of in- travenous lansoprazole administration. The population pharmacokinetics of the drug was evaluated using non- linear mixed effects model (NONMEM) software. In addition, the effect of CYP2C19 polymorphism on proton pump inhibition by lansoprazole was investigated using 24-h intragastric pH monitoring in the 32 subjects. Time course of serum lansoprazole concentration following intravenous short infusion was well described by a 2-com- partment model. The mean volume of the central and peripheral compartments was 0.110 and 0.201 l/kg, respec- tively. The mean inter-compartment clearance was estimated to be 0.0882 l/h/kg. The population mean value of systemic clearance in the homoEM (CYP2C19*1/*1), heteroEM (CYP2C19*1/*2 and *1/*3), and PM (CYP2C19*2/*2, *2/*3, and *3/*3) groups was 0.179, 0.109, and 0.038 l/h/kg, respectively. The mean intragastric pH following twice-daily doses of 30 mg lansoprazole was approximately 6, 5, and 4 in the PM, heteroEM, and homoEM groups, respectively. These findings indicate that large interindividual variability exists in the pharma- cokinetics of intravenously administered lansoprazole, but that twice-daily infusion of a 30 mg dose leads to sig- nificant and sustained proton pump inhibition, even in the homoEM group, despite the short elimination half-life of the drug. Key words lansoprazole; population pharmacokinetics; healthy Japanese subject; CYP2C19

Upper gastrointestinal bleeding is one of the common 2C19.11,12) The genetic polymorphisms of CYP2C19 have causes of emergency hospital admission in Japan, and it pos- been observed in Asian populations, and three polymor- sibly results in life-threatening massive hemorrhage. Endo- phic alleles, CYP2C19*1 (wild-type), CYP2C19*2, and scopic clipping of the bleeding vessel is usually effective for CYP2C19*3, have been identified among Japanese.13) treating gastrointestinal hemorrhage, but rebleeding occurs in CYP2C19*2 causes a splicing defect in exon 5, and 15 to 20% of patients.1) Higher intragastric pH promotes CYP2C19*3 creates a premature stop codon in exon 4.13) platelet aggregation, and the inhibition of secre- Therefore, these mutations lead to the complete loss of the tion to maintain neutral pH may stabilize clots and prevent enzymatic activity of CYP2C19. The allele frequencies of recurrent bleeding.2) Intravenous administration of histamine CYP2C19*2 and CYP2C19*3 among Japanese are reported 14) H2-receptor antagonists has been widely used for the man- to be 28.7% and 13.2%, respectively. It has been reported agement of gastrointestinal hemorrhage, particularly after en- that genetic polymorphisms of CYP2C19 are responsible for doscopic clipping. However, gastric acid secretion is not the large interindividual variability in the pharmacokinetics 15,16) solely regulated by the histamine H2-receptor, and tolerance of orally administered PPIs in Japanese populations. to the antisecretory activity of H2-receptor antagonists fre- Recently, a new preparation for intravenous injection of quently develops during infusion over 72-h periods, leading lansoprazole was developed in Japan. In the present study, to a loss of pH control.1,3) Meta-analyses have also suggested we evaluated the population pharmacokinetics of intra- that H2-receptor antagonists only have a weak beneficial ef- venously administered lansoprazole in healthy Japanese sub- fect.4) On the other hand, proton-pump inhibitors (PPIs) sup- jects using a nonlinear mixed effects model (NONMEM) press gastric acid secretion by specific inhibition of the program. We also evaluated the effect of CYP2C19 polymor- H/K ATPase enzyme system located on the secretory sur- phisms on proton pump inhibition by intravenously adminis- face of gastric parietal cells.5—7) High doses of PPIs maintain tered lansoprazole. the intragastric pH at a nearly neutral level, and inhibit acid production more effectively than an infusion of H2-receptor MATERIALS AND METHODS antagonists.3,8) Therefore, high-dose intravenous PPI therapy is theoretically superior to intravenous H2-receptor antagonist Subjects A total of 56 healthy Japanese adult male vol- therapy for the prevention of recurrent bleeding. In fact, sev- unteers participated in this study. All of the subjects were eral studies have indicated that intravenous administration of confirmed to be healthy on the basis of their medical history, PPIs is effective for the treatment of patients with gastroin- physical findings, laboratory studies, and 12-lead electrocar- testinal bleeding.1,4,9,10) diogram. The subjects ranged in age from 20 to 34 (mean It has been reported that PPIs ( and lansopra- S.D.: 24.04.0) years old, and their weight (S.D.) was zole) are mainly metabolized by cytochrome P450 (CYP) 61.86.0 kg. The subjects were classified into three groups

∗ To whom correspondence should be addressed. e-mail: [email protected] © 2007 Pharmaceutical Society of Japan December 2007 2239 according to their CYP2C19 genotype. The CYP2C19*1 was dissolved in 300 ml of a mixture of water and acetonitrile (wild type) allele and two defective allelic variants, (7 : 3, v/v). A 100 ml aliquot of the solution was injected onto CYP2C19*2 and CYP2C19*3, were determined using the a TSK-GEL ODS 120T column (5 mm, 250 mm4.6 mm allele specific primer polymerase chain reaction (PCR) meth- I.D.). The mobile phase consisted water/acetonitrile/n-octy- od (SNP Typing Kit; Toyobo, Osaka, Japan) or the PCR- lamine (620 : 380 : 1, v/v/v, pH 7). Lansoprazole was detected restriction fragment length polymorphism method.13,17) at 285 nm with an UV detector, and its concentration was de- Sixteen subjects homozygous for the CYP2C19*1 allele were termined from the ratio of peak height of the drug to that of defined as extensive metabolizers, and were designated as isobutyl p-hydroxybenzoate (an internal standard). Linearity homoEM group. Thirty-two subjects heterozygous for mu- was obtained between 5 ng/ml and 2000 ng/ml, and the lower tant alleles (CYP2C19*1/*2 and *1/*3) were defined as inter- limit of quantification was 10 ng/ml for the measurement of mediate metabolizers, and were designated as heteroEM lansoprazole in 500 ml of human serum. group. Finally, 8 subjects homozygous for mutant alleles Population Pharmacokinetic Analysis Population (CYP2C19*2/*2, *2/*3, and *3/*3) were defined as poor me- pharmacokinetic analysis was carried out using the tabolizers, and were designated as PM group. All subjects NONMEM program (version V, level 1.1)19) running on a gave written informed consent to participate in this study, ThinkCentre A50p (8195-CTJ) computer (IBM). We used which was approved by the institutional review boards of the first-order conditional estimation method and NON- Ohsaki Clinic (Tokyo, Japan), Sekino Clinical Pharmacology MEM-PREDPP library subroutines ADVAN3 and TRANS4 Clinic (Tokyo, Japan), and Kyushu Clinical Pharmacology for the 2-compartment model with intravenous dosing. The

Research Clinic (Fukuoka, Japan). volume of the central compartment in the ith individual (V1i) Study Design Serum lansoprazole concentration data for was modeled using the following equation: population pharmacokinetic analysis were obtained during V1i q 1·WT·exp(h V ) (1) the single and multiple dose trials. Briefly, all 56 subjects re- 1i ceived 30 mg lansoprazole by intravenous drip infusion over where q 1 is the predicted population mean volume of the 30 min, and blood samples were collected periodically for the central compartment and h is a random variable with a V1i determination of serum lansoprazole concentration over 12 h mean of zero and variance of w 2 . The clearance in the ith V1i after the dose (see Fig. 1A). To evaluate the constancy of individual (CLi) was modeled using the following equation: pharmacokinetics, 16 (8 PMs, 7 heteroEMs, and 1 homoEM) CLi q 2·q 3·q 4·WT·exp(h CL ) (2) of 56 subjects received twice-daily drip infusions of 30 mg i lansoprazole repetitively for 5 d, and blood samples were col- where q 2 is the predicted population mean systemic clear- lected periodically on day 5 (see Fig. 1B). To assess the lin- ance in the homoEM group, q 3 is the heteroEM/homoEM earity of the pharmacokinetics of lansoprazole, 8 (3 het- clearance ratio, q 4 is the PM/homoEM clearance ratio. That eroEMs, and 5 homoEMs) of 56 subjects received 30 mg of is, both q 3 and q 4 are fixed to one for a subject with lansoprazole as an intravenous bolus injection. Furthermore, CYP2C19*1/*1, q 4 is fixed to one for a subject with 8 (5 heteroEMs, and 3 homoEMs) of 56 subjects received a CYP2C19*1/*2 and *1/*3 alleles, and q 3 is fixed to one for a single dose of 15 mg lansoprazole by drip infusion over 30 subject with CYP2C19*2/*2, *2/*3, and *3/*3. h is a ran- CLi min. Blood samples were collected periodically over 12 h dom variable with a mean value of zero and variance of w 2 . CLi after the dose (see Figs. 1C, D). In addition, the volume of the peripheral compartment in the

Intragastric pH monitoring was performed in 32 of 56 sub- ith individual (V2i) was modeled using the following equa- jects to evaluate the effect of CYP2C19 genotype on proton tion: pump inhibition by 30 mg lansoprazole. In the 16 subjects (8 V2i q 5·WT·exp(h V ) (3) PMs, 7 heteroEMs, and 1 homoEM), who received twice- 2i daily drip infusions of 30 mg lansoprazole repetitively for 5 d, where q 5 is the predicted population mean volume of the pe- 24-h intragastric pH monitoring was performed on day 1 and ripheral compartment and h is a random variable with a V2i day 5. A microglass electrode (CM-181; Chemical Instru- mean of zero and variance of w 2 . Inter-compartmental V2i ments) was introduced per-nasally into the body of the stom- clearance in the ith individual (Qi) was modeled using the ach under fluoroscopic guidance. Intragastric pH data were following equation: recorded with a one-channel pH meter (101ZG; Chemical In- Q q ·WT (4) struments, Tokyo, Japan). For comparison, those 16 subjects i 6 also received twice-daily drip infusions of 75 mg roxatidine where q 6 is the predicted population mean value for inter- acetate repetitively for 5 d, and 24-h intragastric pH monitor- compartmental clearance. Finally, the serum lansoprazole ing was performed on day 1 and day 5 as described above. In concentration was modeled using the following equation: additional 16 subjects (8 heteroEMs, and 8 homoEMs), 30 Cij Cpred,ij·(1 e CV ) e ADD (5) mg lansoprazole was infused intravenously two times a day ij ij at a 12-h interval, and 24-h intragastric pH monitoring was where Cij is the jth observed serum concentration in the ith performed on the day (day 1), as described above. subject, Cpred,ij is the jth predicted serum concentration in the Assay of Lansoprazole The serum concentration of lan- ith subject, and e and e are residual intraindividual CVij ADDij 2 soprazole was determined by the HPLC method reported by variations with a mean value of zero and variances of s CV Aoki et al.18) Briefly, lansoprazole was extracted from 500 ml and s 2 . The magnitude of the variation of e can be ADD CVij of human serum with 3 ml of a mixture of diethyl ether and viewed as the ‘proportional’ component of the residual error dichloromethane (7 : 3, v/v). The extract was evaporated to model, while that of e can be seen as the ‘additive’ com- ADDij dryness under a stream of nitrogen gas, and then the residue ponent. NONMEM provides estimates of the standard error 2240 Vol. 30, No. 12

(S.E.) for all parameters, and S.E. data can be used to define was 9402.358, whereas those for the reduced models with 95% confidence intervals (CI) for true parameter values as q 3 1 and q 3 q 4 1 were 9470.184 and 9544.402, respec- follows: 95% CI(estimated parameter value)1.96·S.E.19) tively. The observed plasma lansoprazole concentrations in 56 subjects were also compared with the predicted concen- RESULTS AND DISCUSSION trations in order to evaluate the reliability of the population pharmacokinetic parameters (Fig 2). The drug concentration

The main objective of the present study was to estimate in Fig. 2A was predicted by population parameters (q 1–q 6) the population pharmacokinetic parameters of intravenously shown in Table 1, whereas that in Fig. 2B was predicted by administered lansoprazole, and also to evaluate the effect of the parameter set obtained with q 3 q 4 1. The correlation CYP2C19 polymorphisms on the pharmacokinetics of the between the predicted and observed plasma concentrations drug. The pharmacokinetic data were obtained from a total was fairly good in Fig. 2A, but not in Fig. 2B. These findings of 56 subjects (8 PMs, 32 heteroEMs, and 16 homoEMs). indicated that the significant effect of CYP2C19 defects on The population pharmacokinetic analysis was performed the systemic clearance of lansoprazole. using NONMEM software, because the NONMEM analysis A 1-compartment model analysis was performed to evalu- enables us to simultaneously evaluate the mean pharmacoki- ate the validity of the 2-compartment model (Table 1). The netic parameters, the covariates affecting the pharmacokinet- mean value of CL in the homoEM group (q 2) was estimated ics of a drug, and also the intra- and interindividual variabil- to be 0.169 l/h/kg for the 1-compartment model. The mean ity of the pharmacokinetics. In addition, the effect of values of CL in the heteroEM and PM groups were estimated

CYP2C19 polymorphisms on proton pump inhibition by lan- to be 60.9% (q 3) and 21.7% (q 4) of that in the homoEM soprazole was evaluated with 24-h intragastric pH monitor- group, respectively. In addition, w CL value was estimated to ing. The pharmacodynamic data were obtained from 32 (8 be 0.208. These values were comparable with those obtained PMs, 15 heteroEMs, and 9 homoEMs) of 56 subjects. from the 2-compartment model analysis (Table 1). However,

Figure 1 shows the observed serum concentrations of lan- the s CV value for the 1-compartment model analysis (0.223) soprazole following intravenous administration. Marked in- was much greater than that for the 2-compartment model terindividual variability was observed in the serum concen- analysis (0.0604). In addition, the minimal objective function tration profiles following intravenous infusion of 30 mg lan- (2log likelihood) value for a 1-compartment model soprazole over 30 min (Fig. 1A). Higher lansoprazole levels (114473.362), which was produced by NONMEM software, were observed in the heteroEM group when compared to the was much greater than that obtained from the 2-compartment homoEM group. In addition, the serum lansoprazole levels in model analysis (9402.358). These results indicate that the the PM group were considerably higher than those in the het- time course of serum lansoprazole concentration following eroEM group. The elimination half-life of lansoprazole fol- intravenous administration can be better described by a 2- lowing single intravenous administration was significantly compartment model. Furthermore, Fig. 3 shows the predicted shorter in the homoEM and heteroEM groups than in the PM serum concentration profile of lansoprazole in a typical 60- group (Fig. 1A). Concentration–time curves for lansoprazole kg subject who received twice-daily infusions at the dose of following repetitive twice-daily infusions at 5 d were similar 30 mg for 5 d. The peak lansoprazole concentrations simu- to those on day 1 (Figs. 1A, B). The distribution and elimina- lated with the 1-compartment model were much lower than tion half-life of lansoprazole after bolus injection at a dose of those simulated with the 2-compartment model, indicating 30 mg was similar to that after the 30-min infusion, and the insufficiency of 1-compartment model analysis. serum drug concentrations in the heteroEM group were also Figure 4 shows the individual clearance values of lanso- higher than those in the homoEM group (Figs. 1A, C). In ad- prazole in 3 genotype groups, where the pharmacokinetic pa- dition, the serum concentrations of lansoprazole after a sin- rameters for a 2-compartment model in 56 individual sub- gle infusion of 15 mg were approximately 50% lower than jects were obtained from population estimates according to those after 30 mg infusion (Figs. 1A, D). These findings indi- Baye’s theorem using the NONMEM posthoc option. The ef- cate that the pharmacokinetics of lansoprazole administered fect of CYP2C19*2 and *3 on individual clearance values intravenously was constant and linear in the tested dose was significant, and the mean clearance values in 16 ho- range. moEMs, 32 heteroEMs, and 8 PMs were 0.187, 0.109, and One thousand and sixty-nine of serum lansoprazole con- 0.039 l/h/kg, respectively. These mean clearance values coin- centration data points from a total of 56 subjects were ana- cided very well with those calculated from the population lyzed using a 2-compartment pharmacokinetic model. The pharmacokinetic parameters in Table 1 (0.179, 0.179·0.612, population pharmacokinetic parameters are summarized in and 0.179·0.212 l/h/kg). The gene-dose effect of CYP2C19

Table 1. The mean value of V1 (q 1) was estimated to be polymorphisms on the clearance of lansoprazole seemed to 0.110 l/kg, and the mean value of CL in the homoEM group be comparable with that of intravenously infused omepra- 15) (q 2) was estimated to be 0.179 l/h/kg. The mean value of CL zole. Specifically, Uno et al. reported that the mean clear- in the heteroEM group was estimated to be 61.2% (q 3) of ance values of omeprazole given intravenously at the dose of that in the homoEM group. The mean value of CL in the PM 20 mg were 14.6, 8.9, and 3.0 l/h in homoEMs, heteroEMs, 15) group was estimated to be only 21.2% (q 4) of that in the ho- and PMs, respectively. moEM group. The mean values of V2 (q 5) and Q (q 6) were In the present study, 16 subjects (8 PMs, 7 heteroEMs, and estimated to be 0.201 l/kg and 0.0882 l/h/kg, respectively. 1 homoEM) received twice-daily drip infusions of 75 mg

The w V1, w CL, and q V2 values for a 2-compartment model repetitively for 5 d, and 24-h intragastric were estimated to be 0.149, 0.221, and 0.0880, respectively pH monitoring was performed on day 1 and day 5. Figure 5A (Table 1). The objective function value for this full model shows the mean 24-h intragastric pH profiles before and after December 2007 2241

Fig. 1. Serum Concentration of Lansoprazole Following Intravenous Administration Lansoprazole was administered by a single infusion of 30 mg over 30 min (A), twice-daily infusion of 30 mg over 30 min for 5 d (B), a bolus injection of 30 mg (C), and a single infusion of 15 mg over 30 min (D). Serum concentrations of lansoprazole in the homoEM, heteroEM, and PM groups are colored with blue, green, and red, respectively.

Fig. 2. The Relationship between the Predicted and Observed Lansoprazole Concentrations in 56 Subjects

The drug concentration in Fig. 2A was predicted by population parameters (q 1–q 6) shown in Table 1, whereas that in Fig. 2B was predicted by the parameter set obtained with q 3 q 4 1.

Table1.Population Pharmacokinetic Parameters of Intravenously Admin- istered Lansoprazole in Healthy Japanese Subjects

2-Compartment 1-Compartment Parameter Estimate 95% CI Estimate 95% CI

q 1 (l/kg) 0.110 0.104–0.116 0.185 0.181–0.189 q 2 (l/h/kg) 0.179 0.163–0.195 0.169 0.154–0.184 q 3 0.612 0.548–0.676 0.609 0.547–0.671 q 4 0.212 0.178–0.246 0.217 0.184–0.250 q 5 (l/kg) 0.201 0.179–0.223 q 6 (l/h/kg) 0.0882 0.0838–0.0926 w 0.149 0.126–0.169 0.0829 0.0643–0.0981 V1 w CL 0.221 0.191–0.247 0.208 0.176–0.235 w 0.0880 0.0525–0.113 V2 s CV 0.0604 0.0521–0.0677 0.223 0.209–0.236 s ADD (ng/ml) 6.40 0.00–10.2 6.90 2.50–9.45 administration of roxatidine acetate. The mean intragastric pH in the 16 subjects before administration of roxatidine ac- etate was approximately 2, and the pH value was transiently increased by food intake. Because roxatidine acetate is pre- 20) dominantly eliminated by renal , H2-receptor blockade by the drug was not influenced by the CYP2C19 Fig. 3. Serum Concentration Profile of Lansoprazole during Twice-Daily polymorphisms in the present study. The intragastric pH on Intravenous Administration of 30 mg for 5 d day 1 was increased by treatment with 75 mg roxatidine ac- Lines are simulation curves obtained from pharmacokinetic analysis with a 2-com- partment model (A) and a 1-compartment model (B). Serum concentrations of lanso- etate (Fig. 5A). The percentage of time with intragastric pH prazole in the homoEM, heteroEM, and PM groups are colored with blue, green, and less than 4 is highly correlated with severity of mucosal in- red, respectively. 2242 Vol. 30, No. 12

jury, and drug therapy that raises intragastric and/or intrae- sophageal pH above 4 achieves the best clinical results.21,22) The inhibitory effect of roxatidine acetate on gastric acid se- cretion, therefore, was evaluated as the percentage of time with intragastric pH above 4. The mean (S.D.) percent of time with a pH above 4 was 43.918.7% and 30.315.8% in the daytime (9:00–21:00) and nighttime (21:00–9:00), re- spectively. On the other hand, the intragastric pH on day 5 was lower than that on day 1 (Fig. 5A), suggesting that par- tial tolerance to the antisecretory effect of roxatidine acetate developed following repetitive administration of the drug. This attenuation of antisecretory activity has also been docu- 3,23) mented for other H2-receptor antagonists. Netzer et al. re- ported that tolerance to the antisecretory activity of Fig. 4. Effect of CYP2C19 Polymorphisms on the Individual Clearance of Lansoprazole develops during infusion for 72-h periods leading to a loss of 3) Circles correspond to individual estimates. Horizontal bars represent the meanS.D. pH control. In addition, Komazawa et al. reported that tol- for each genotype group. erance occurs during 14 d of continuous adminis- tration.23) In the present study, 16 subjects (8 PMs, 7 heteroEMs, and 1 homoEM) also received 30 mg lansoprazole repetitively for 5d, and 24-h intragastric pH monitoring was performed on day 1 and day 5. In additional 16 subjects (8 heteroEMs, and 8 homoEM), 30 mg lansoprazole was infused two times at a 12-h interval, and 24-h intragastric pH monitoring was per- formed on day 1. Figure 5B shows the mean 24-h intragastric pH profile before and after administration of 30 mg lansopra- zole to 8 PMs. The mean value of intragastric pH on day 1 was approximately 6, and the mean (S.D.) percent of time with pH above 4 was 84.35.5% and 95.96.1% in the day- time and nighttime, respectively (Fig. 5B). Figure 5C shows the mean 24-h intragastric pH profiles before and after ad- ministration of 30 mg lansoprazole to 15 heteroEMs. The mean value of intragastric pH on day 1 was approximately 5, and the mean (S.D.) percent of time with pH above 4 was 70.716.2% and 79.214.0% in the daytime and nighttime, respectively (Fig. 5C). Figure 5D shows the mean 24-h intra- gastric pH profile before and after administration of 30 mg lansoprazole to 9 homoEMs. The mean value of intragastric pH on day 1 was approximately 4, and the mean (S.D.) per- cent of time with pH above 4 was 43.930.7% and 53.5 30.4% in the daytime and nighttime, respectively (Fig. 5D). These results indicate that twice-daily infusion of a 30 mg dose leads to significant proton pump inhibition even in ho- moEMs despite the short elimination half-life of the drug (Fig. 1). The present findings for lansoprazole were compara- ble to earlier findings for omeprazole.24) Sugimoto et al. pre- viously reported that the median intragastric pH for the first 24 h following twice daily intravenous administration of 20 mg omeprazole was 6.1, 5.8 and 3.9 in PM, heteroEM, and homoEM groups, respectively.24) In addition, no significant tolerance to the antisecretory effect of lansoprazole was ob- served in the present study. Although the number of subjects who received twice-daily infusions of 30 mg lansoprazole for 5d was only 8, 7, and 1 in the PM (Fig. 5B), heteroEM (Fig. Fig. 5. Mean Intragastric pH Profile Following Intravenous Administra- 5C), and homoEM (Fig. 5D) groups, respectively, the mean tion of Roxatidine Acetate and Lansoprazole intragastric pH in these subjects on day 5 was still higher A dose of 75 mg roxatidine acetate was given twice-daily for 16 subjects (A). A dose than that on day 1. of 30 mg lansoprazole was given twice-daily to 8 PMs (B), 15 heteroEMs (C), and 9 homoEMs (D). Mean intragastric pH during the baseline period, at day 1, and at day 5 In conclusion, we estimated the population pharmacoki- are depicted with a thin solid line, thick solid line, and dotted line, respectively. netic parameters of lansoprazole using NONMEM software and evaluated proton pump inhibition in healthy male Japan- ese subjects. The present study indicates that CYP2C19*2 December 2007 2243 and *3 alleles are responsible for the pharmacokinetic vari- Gut, 49, 372—379 (2001). ability of intravenously administered lansoprazole among 11) Furuta T., Shirai N., Sugimoto M., Nakamura A, Hishida A., Ishizaki T., Drug Metab. Pharmacokinet., 20, 153—167 (2005). Japanese. These findings may provide information for the 12) Pearce R. E., Rodrigues A. D., Goldstein J. A., Parkinson A., J. Phar- proper use of lansoprazole injections to treat patients with macol. Exp. Ther., 277, 805—816 (1996). upper gastrointestinal bleeding. 13) de Morais S. M. F., Wilkinson G. R., Blaisdell J., Meyer U. A., Naka- mura K., Goldstein J. A., Mol. Pharmacol., 46, 594—598 (1994). 14) Kubota T., Chiba K., Ishizaki T., Clin. Pharmacol. Ther., 60, 661—666 REFERENCES (1996). 15) Uno T., Niioka T., Hayakari M., Yasui-Furukori N., Sugawara K., 1) Lau J. Y., Sung J. J., Lee K. K., Yung M. Y., Wong S. K., Wu J. C., Tateishi T., Eur. J. Clin. Pharmacol., 63, 143—149 (2007). Chan F. K., Ng E. K., You J. H., Lee C. W., Chan A. C., Chung S. C., 16) Sakai T., Aoyama N., Kita T., Sakaeda T., Nishiguchi K., Nishitora Y., N. Engl. J. Med., 343, 310—306 (2000). Hohda T., Sirasaka D., Tamura T., Tanigawara Y., Kasuga M., Oku- 2) Green F. W., Jr., Kaplan M. M., Curtis L. E., Levine P. H., Gastroen- mura K., Pharm. Res., 18, 721—727 (2001). terology, 74, 38—43 (1978). 17) de Morais S. M. F., Wilkinson G. R., Blaisdell J., Nakamura K., Meyer 3) Netzer P., Gaia C., Sandoz M., Huluk T., Gut A., Halter F., Husler J., U. A., Goldstein J. A., J. Biol. Chem., 269, 15419—15422 (1994). Inauen W., Am. J. Gastroenterol., 94, 351—357 (1999). 18) Aoki I., Okumura M., Yashiki T., J. Chromatogr., 571, 283—290 4) Huggins R. M., Scates A. C., Latour J. K., Ann. Pharmacother., 37, (1991). 433—437 (2003). 19) Beal S. L., Boeckmann A. J., Sheiner L. B., “NONMEM Users 5) Kubo K., Oda K., Kaneko T., Satoh H., Nohara A., Chem. Pharm. Guides: NONMEM Project Group,” University of California, San Bull., 38, 2853—2858 (1990). Francisco, 1992. 6) Satoh H., Inatomi N., Nagaya H., Inada I., Nohara A., Nakamura N., 20) Collins J. D., Pidgen A. W., Drugs, 35, 41—47 (1988). Maki Y., J. Pharmacol. Exp. Ther., 248, 806—815 (1989). 21) Bell N. J., Burget D., Howden C. W., Wilkinson J., Hunt R. H., Diges- 7) Nagaya H., Satoh H., Maki Y., J. Pharmacol. Exp. Ther., 252, 1289— tion, 51 (Suppl. 1), 59—67 (1992). 1295 (1990). 22) Hunt R. H., Arch. Intern. Med., 159, 649—657 (1999). 8) Laine L., Gastrointest. Endosc., 57, 557—560 (2003). 23) Komazawa Y., Adachi K., Mihara T., Ono M., Kawamura A., Fujishiro 9) Leontiadis G. I., Sharma V. K., Howden C. W., Br. Med. J., 330, 568— H., Kinoshita Y., J. Gastroenterol. Hepatol., 18, 678—682 (2003). 570 (2005). 24) Sugimoto M., Furuta T., Shirai N., Ikuma M., Hishida A., Ishizaki T., 10) Hawkey G. M., Cole A. T., McIntyre A. S., Long R. G., Hawkey C. J., Clin. Pharmacol. Ther., 80, 539—548 (2006).