Proton pump inhibitors: an update of their clinical use and pharmacokinetics

Transcription

1 Eur J Clin Pharmacol (2008) 64: DOI /s y REVIEW ARTICLE Proton pump inhibitors: an update of their clinical use and pharmacokinetics Shaojun Shi & Ulrich Klotz Received: 21 January 2008 /Accepted: 1 July 2008 /Published online: 5 August 2008 # Springer-Verlag 2008 Abstract Background Proton pump inhibitors (PPIs) represent drugs of first choice for treating peptic ulcer, Helicobacter pylori infection, gastrooesophageal reflux disease, nonsteroidal anti-inflammatory drug (NSAID)-induced gastrointestinal lesions (complications), and Zollinger-Ellison syndrome. Results The available agents (omeprazole/esomeprazole, lansoprazole, pantoprazole, and rabeprazole) differ somewhat in their pharmacokinetic properties (e.g., time-/dose-dependent bioavailability, metabolic pattern, interaction potential, genetic variability). For all PPIs, there is a clear relationship between drug exposure (area under the plasma concentration/ time curve) and the pharmacodynamic response (inhibition of acid secretion). Furthermore, clinical outcome (e.g., healing and eradication rates) depends on maintaining intragastric ph values above certain threshold levels. Thus, any changes in drug disposition will subsequently be translated directly into clinical efficiency so that extensive metabolizers of CYP2C19 will demonstrate a higher rate of therapeutic nonresponse. Conclusions This update of pharmacokinetic, pharmacodynamic, and clinical data will provide the necessary guide by S. Shi : U. Klotz (*) Dr. Margarete Fischer-Bosch-Institut für Klinische Pharmakologie, Auerbachstraße 112, Stuttgart, Germany ulrich.klotz@ikp-stuttgart.de S. Shi : U. Klotz University of Tuebingen, Tuebingen, Germany Present address: S. Shi Department of Pharmacy of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, People s Republic of China which to select between the various PPIs that differ based on pharmacodynamic assessments in their relative potencies (e.g., higher doses are needed for pantoprazole and lansoprazole compared with rabeprazole). Despite their well-documented clinical efficacy and safety, there is still a certain number of patients who are refractory to treatment with PPIs (nonresponder), which will leave sufficient space for future drug development and clinical research. Keywords Proton pump inhibitors. Pharmacokinetics. Pharmacodynamics. Peptic ulcer. Reflux disease. Helicobacter pylori infection Introduction Proton pump inhibitors (PPIs) are the mainstays in treating acid-related diseases. Since the introduction of omeprazole in 1989, other PPIs became available, e.g., lansoprazole (1995), pantoprazole (1997), rabeprazole (1999), and the S- enantiomer of omeprazole (2001). PPIs inhibit selectively and irreversibly the gastric H + /K + ATPase (the proton pump) that accomplishes the final step in acid secretion. All PPIs inhibit both basal and stimulated secretion of gastric acid, independent of the nature of parietal cell stimulation [1]. PPIs undergo extensive hepatic metabolism by the cytochrome P450 (CYP) system, and CYP2C19 polymorphisms have been shown to substantially influence the pharmacokinetics, pharmacodynamics, and clinical outcome of PPIs [2, 3]. In addition, pharmacokinetic and pharmacodynamic differences between PPIs are reflected in their influence on both speed and degree of gastric acid suppression, which subsequently may affect their clinical efficacy [4].

2 936 Eur J Clin Pharmacol (2008) 64: PPIs are drugs of first choice for peptic ulcers (PU) and their complications (e.g., bleeding), gastrooesophageal reflux disease (GERD), nonsteroidal anti-inflammatory drug (NSAID)-induced gastrointestinal (GI) lesions, Zollinger-Ellison syndrome, dyspepsia, and eradication of Helicobacter pylori (H. pylori) with two antibiotics [5, 6]. PPIs have demonstrated an excellent safety profile after approximately two decades of clinical use [7]. There are some unspecific adverse events (AEs) such as headache, nausea, and diarrhea, but the main concern is emerging from the long-term suppression of acid secretion [8]. During the last few years, much new data have been published on the pharmacological characteristics and therapeutic efficacy of PPIs. Therefore, it appears appropriate to provide an update on the pharmacokinetics, pharmacodynamic action, and clinical use of PPIs. Pharmacokinetic properties PPIs are substituted benzimidazole derivatives and membrane-permeable, weak bases that accumulate in acid spaces of the active parietal cell as prodrugs. Here, they undergo acid-catalyzed conversion to the active sulfenic acid and sulfonamide derivatives. These bind covalently via disulfide bridges to cysteine residues on the alpha subunit of the H + /K + ATPase, thus inhibiting acid secretion up to 36 h [1]. Whereas racemic PPI prodrugs possess a chiral center, the identical biological active principle (which is optically inactive) is formed from both enantiomers. The main pharmacokinetic parameters of all PPIs are compared in Table 1 [6, 9]. Omeprazole and esomeprazole Omeprazole is administered as a racemic mixture of its two enantiomers, S-omeprazole (esomeprazole) and R-omeprazole. Both prodrugs are acid labile and usually administered as encapsulated enteric-coated granules. After oral administration, they are rapidly absorbed [9]. They should be swallowed at least 1 h before eating. The bioavailability (F) of esomeprazole was significantly reduced when taken within 15 min before eating a high-fat meal compared with fasting conditions [10]. When increasing the dose and after repeated administration of omeprazole or esomeprazole, the maximal plasma concentration (C max ) and area under the concentration-time curve (AUC) increased in a nonlinear fashion [11], which is due to decreased first-pass elimination and decreased systemic clearance (CL). Furthermore, due to a lower metabolic rate of esomeprazole compared with R- or racemic omeprazole, esomeprazole resulted in higher AUC values [11]. Similarly, in patients with GERD, the C max and AUC values of esomeprazole increased on day 7 versus day 1 by 80% and 50%, respectively [12]. Omeprazole and esomeprazole are extensively metabolized by CYP2C19 and CYP3A4. Omeprazole is converted mainly to hydroxyl and 5-O-desmethyl metabolites by CYP2C19 and to the sulfone by CYP3A4 [6]. In vitro studies suggest that esomeprazole is predominantly metabolized by CYP3A4 and consequently is less dependent on CYP2C19, which recently could be substantiated by a clinical study [13]. Approximately 80% of each dose is excreted as metabolites in the urine [6]. CYP2C19 polymorphisms can significantly influence the metabolism of omeprazole and esomeprazole [3, 14]. Table 1 Pharmacokinetic properties of proton pump inhibitors (according to Klotz [6]) Parameter Omeprazole Esomeprazole Lansoprazole Pantoprazole Rabeprazole t max [h] F [%] ( upon multiple 50 (acute dosing) dosing) (chronic dosing) Linear pharmacokinetics No No Yes Yes Yes fu [%] V [l/kg] t 1/2 [h] CL [ml/min] (acute) (chronic) CL/F [ml/min] f e [%] Negligible Negligible Negligible Negligible Negligible Effect of age CL, t 1/2 ( ), F CL, t 1/2 CL, t 1/2 CL, t 1/2 CL, t 1/2 ( ) Renal insufficiency CL, t 1/2 ( ), F CL, t 1/2 ( ) CL, t 1/2 CL( ), t 1/2 Hepatic dysfunction CL, t 1/2,F CL, t 1/2 CL, t 1/2 CL, t 1/2,F CL, t 1/2 t max time to maximal plasma concentration, F oral bioavailability, fu fraction of drug unbound in plasma, V apparent volume of distribution, t 1/2 elimination half life, CL systemic clearance, CL/F apparent oral clearance, f e fraction excreted in unchanged form into urine increase, decrease, no significant change, arrows in parentheses effects are equivocal

3 Eur J Clin Pharmacol (2008) 64: According to metabolic rate (phenotype), individuals can be classified as homozygous extensive metabolizers (homem), heterozygous extensive metabolizers (hetem) and poor metabolizers (PM). The frequencies of these three subgroups show a wide interethnic variation. The prevalence of PM ranges from 1.2% to 3.8% in Caucasian Europeans and up to 23% in Asian Oceanian populations [15, 16]. PMs exhibit a 3- to 10-fold and hetems a 2- to 3-fold higher AUC (drug exposure) compared with homems [3, 14]. In a recent study, mean AUC ratios of omeprazole were 1:2.7:9.0 after a single oral dose of 40 mg and 1:1.7:4.3 after a single i.v. dose of 20 mg in six homems, eight hetems, and six PMs, respectively. Therefore, in PMs, F was higher than that in homems and hetems (87% vs. 62% and 41%; P<0.001) [17]. Similarly, in homems, hetems, and PMs, the relative AUC ratios were 1:2.8:7.5 and t 1/2 ratios 1:1:1.7 for omeprazole, indicating that disposition of omeprazole is greatly dependent on the CYP2C19 genotype [18]. Similar ratios were reported after repeated doses of omeprazole [19]. Interestingly, the elderly EMs showed a wide variance in their CYP2C19 activity and were phenotypically closer to the PMs than the young EMs compared with the young PMs. Thus, in the elderly, the CYP2C19 genotype may not be as useful as phenotyping [20]. The most frequently and extensively described variant alleles for PMs are CYP2C19*2 and CYP2C19*3, which encode for nonfunctional proteins. Recently, the CYP2C19*17 allele has been identified, which is associated with a very rapid metabolism phenotype and has a frequency of 18% both in Swedes and Ethiopians but only 4% in Chines populations [21]. Such subjects need higher doses of omeprazole for acid suppression [22, 23]. The pharmacokinetics of omeprazole was compared in 18 healthy adults and in 12 children with GERD (mean age 6.1 years). Oral clearance (CL/F) and apparent volume of distribution (V) in healthy adults (0.62 L/h per kg and 0.76 L/kg, respectively) were not significantly different from those in children with GERD (0.51 L/h per kg and 0.66 L/kg, respectively). Therefore, dosing on a milligram/ kilogram basis was recommended, with no further adjustments for treating GERD in children [24]. In 27 children with GERD aged 1 11 years, the pharmacokinetic properties of esomeprazole were both dose (between 5 20 mg) and age dependent. The younger children (1 5 years) showed a more rapid metabolism compared with older children (6 11 years) [25]. In 28 adolescent patients with GERD aged years, the mean AUC and C max values of esomeprazole were 3.5-fold higher with the 40 mg dose compared with the 20 mg dose with single- and repeated-dose administration, confirming nonlinear pharmacokinetics [26]. All PPIs cause increases in intragastric ph, which can affect absorption of concomitantly given drugs, such as ketoconazole, vitamin B 12, and digoxin [27]. In addition, omeprazole carries the potential for inhibiting the hepatic elimination of a wide range of drugs, including carbamazepine, diazepam, mephenytoin, methotrexate, nifedipine, phenytoin, warfarin, mefloquine, pyrimethamine, and sulfadoxine [6, 27, 28]. Likewise, inhibitors of CYP2C19 or CYP3A4, e.g., fluconazole [29] and fluvoxamine [30], can affect the metabolism of omeprazole. Furthermore, omeprazole is a substrate and inhibitor of P-glycoprotein. So drug interactions could be also partly due to an inhibition of P- glycoprotein-mediated drug transport [31, 32]. On the contrary, St. John s wort can induce both pathways in the metabolism of omeprazole and decreased its AUC (38 44%) and C max (38 50%) [33]. As with smoking, omeprazole can dose dependently induce CYP1A2 [6]. In general, it can be expected that the interaction potential of esomeprazole is similar to that of racemic omeprazole [27, 34]. Triple therapy with omeprazole/clarithromycin/amoxicillin is widely used to eradicate H. pylori. The observed pharmacodynamic synergism of these drugs is at least partly due to pharmacokinetic interactions. The AUC of omeprazole increased almost twofold after concomitant administration of clarithromycin [35]. Naproxen and rofecoxib did not interact with esomeprazole and visa versa. Apparently, esomeprazole can be used in combination with NSAIDs without the risk of pharmacokinetic interactions [36]. Lansoprazole Lansoprazole is rapidly absorbed and displays a linear increase in plasma concentrations over a dose range of mg [37]. Pharmacokinetics of repeated doses is similar to that of a single dose. Lansoprazole is extensively and rapidly (t 1/2 : 1 2 h) metabolized into sulfone and 5- hydroxylated metabolites by CYP3A4 and CYP2C19. A sulfide metabolite is also present in smaller amounts [6, 37]. As CYP2C19 is involved in the metabolism of lansoprazole, it is not surprising that PMs have a 4.4-fold higher AUC than homems [3]. The pharmacokinetics of lansoprazole (15 mg/day) in children with GERD aged months was comparable to those in older children and adults [38]. Fluvoxamine treatment increased AUC of lansoprazole 3.8-fold (P<0.01) in homems and 2.5-fold (P<0.05) in hetems, whereas in PMs, no difference was found in any pharmacokinetic parameter [39]. A double-blind, randomized controlled trial (RCT) showed that clarithromycin treatment significantly increased C max of lansoprazole 1.47-, 1.71-, and 1.52-fold and AUC 1.55-, 1.74-, and 1.80-fold in homems, hetems, and PMs, respectively. These very similar effects indicate that the drug interactions resulted from inhibition of CYP3A4 [40]. Surprisingly, grapefruit juice had no significant effect on the pharmacokinetics of lansoprazole [41].

4 938 Eur J Clin Pharmacol (2008) 64: Lansoprazole did not demonstrate relevant interactions with theophylline, phenytoin, prednisone, warfarin, diazepam, oral contraceptives, ivabradine, or methotrexate [37, 42, 43]. In renal transplant recipients receiving tacrolimus and lansoprazole (30 mg) or rabeprazole (20 mg), elevated blood concentrations of tacrolimus were observed only in the very rare subgroup of subjects who are PM of CYP2C19 and bearing also the CYP3A5*3/*3 genotype [44]. Absorption of atazanavir was significantly reduced when coadministered with lansoprazole, as evidenced by a 94% decline in AUC and by a 96% decrease in C max [45]. Pantoprazole Pantoprazole is rapidly absorbed after oral administration of enteric-coated tablets, which avoid degradation of the PPI by gastric acid [6]. It hardly undergoes first-pass metabolism and has a bioavailability of approximately 77%, independent of dose and food intake. Pantoprazole shows linear pharmacokinetics after both i.v. and oral administration. It is completely metabolized by CYP2C19 and CYP3A4. In a major pathway, pantoprazole undergoes O- demethylation followed by sulfate conjugation and sulfone/ sulfide formation [46]. Pantoprazole has apparently no clinically relevant drug interactions, other than the class effect associated with elevated intragastric ph. No drug interactions have been demonstrated between pantoprazole and a wide range of drugs, such as theophylline, diazepam, carbamazepine, digoxin, and warfarin [46, 47]. Unlike omeprazole, administration of clarithromycin did not increase pantoprazole levels [35]. Rabeprazole In healthy subjects, mean F has been calculated to 52% [48]. C max and AUC values were linearly related to the dose, whereas t max and t 1/2 were dose independent. Rabeprazole s metabolism is unique because of its reduction to rabeprazole thioether via a nonenzymatic pathway with renal elimination of the metabolites. Both CYP2C19 and CYP3A4 contribute only a small fraction to the overall metabolism [49]. Thus, rabeprazole will be less susceptible to the influence of genetic polymorphisms of either CYP2C19 or CYP3A4 [50]. In healthy Chinese subjects, the pharmacokinetics of rabeprazole was dependent to a certain degree on the CYP2C19 genotype. AUC values for rabeprazole differed among the three genotype groups (homem, hetem, PM), with relative ratios of 1.0:1.3:1.8 after a single dose and of 1.0:1.1:1.7 after repeated doses, respectively. These changes were much smaller than those observed for other PPIs [51]. The modest differences in AUC of rabeprazole were also consistent with other studies [18, 52]. However, in 12 Chinese subjects receiving rabeprazole 20 mg b.i.d., the mean AUC values of rabeprazole and rabeprazole thioether were higher (P< 0.001) in PMs than in EMs on day 1. Similar results were observed on day 4, which might be due to the high rabeprazole dose (40 mg/day) used in this study [53]. Pharmacokinetic properties of rabeprazole in children years old were similar to adults, and during multiple dosing, only headache and nausea were sometimes reported [54]. As the CYP450-mediated pathways are secondary in rabeprazole metabolism, it has a low potential for drug interactions involving CYPs. Interaction studies with rabeprazole revealed no significant metabolic effect on theophylline, phenytoin, warfarin, or diazepam [27, 49]. Clarithromycin or verapamil did not alter the pharmacokinetics of rabeprazole, irrespective of the CYP2C19 genotypes [55]. Interestingly, fluvoxamine, an inhibitor of CYP1A2 and CYP2C19, increased AUC of rabeprazole and rabeprazole thioether 2.8- and 5.1-fold in homems and 1.7- and 2.6-fold in hetems, respectively, whereas no difference was seen in PMs of CYP2C19 [56]. According to in vitro experiments all PPIs showed some inhibition of CYP2C9, 2C19, and 3A4; the inhibitory potency of rabeprazole was relatively lower than that of the other PPIs [57]. Pharmacodynamic profile The primary effect of PPIs is suppression of gastric acid secretion, which will determine healing rates in GERD and PU [58]. Intragastric ph monitoring allows direct assessment of acid suppression, and it is a very useful measure with which to compare antisecretory therapies [59]. Typically, the antisecretory effects of PPIs have been reported as mean or median 24-h intragastric ph or suppression of 24-h intragastric acidity expressed as time above a predefined ph threshold. Intragastric ph value above 3 (PU) and 4 (GERD) have been defined as therapeutic targets [60]. PPIs are regarded to be similarly effective, but their potency differs. Such differences may translate into a slight advantage for a particular PPI in a special clinical situation [61]. Onset and degree of acid suppression All PPI prodrugs undergo accumulation and acid activation in the parietal cell, which relies on their pk a values. Based on in vitro experiments and pk a considerations, it can be estimated that rabeprazole will show the highest accumulation and faster conversion to the active form of the PPIs [1]. This might affect the speed of acid suppression by PPIs. Apparently an earlier sustained inhibition of acid

5 Eur J Clin Pharmacol (2008) 64: secretion was seen with rabeprazole than with other PPIs [62]. Comparing the antisecretory effects on the first day of dosing, rabeprazole achieved better acid control because the intragastric ph (3.4) and the time of ph >4 during the 24-h postdose were significantly (P 0.04) greater with rabeprazole than with lansoprazole, pantoprazole, or two formulations of omeprazole [63]. In contrast, in 72 healthy volunteers, mean 24-h ph and percentage of time for ph>4 were not significantly different between lansoprazole 30 mg and rabeprazole 20 mg. Lansoprazole resulted in greater acid suppression during hours 0 5 on days 1 and 5, whereas rabeprazole had greater suppression during hours on day 5 [64]. In patients with heartburn, the intragastric ph profiles of all five PPIs were compared after giving once daily esomeprazole 40 mg, lansoprazole 30 mg, omeprazole 20 mg, pantoprazole 40 mg, and rabeprazole 20 mg [65, 66]. The results of day 5 suggest that intragastric ph control with esomeprazole 40 mg/day was superior to other PPIs [65]. Similarly, with the same dosing schedule, esomeprazole maintained intragastric ph >4 for a longer time compared with all other PPIs on days 1 and 5 [66]. In seven healthy homems, the median intragastric ph and percent time ph>4 for 24 h increased dose dependently with omeprazole 10, 20, and 40 mg daily for 7 days, and 10 and 20 mg b.i.d. were comparable with once-daily 20 and 40 mg. Concerning percent time ph>4 at nighttime, omeprazole 20 mg b.i.d. was superior (P<0.05) to 40 mg daily [67]. Three RCTs indicated that a single dose of esomeprazole (40 mg i.v.) was superior to omeprazole (40 mg i.v.) in reducing stimulated acid secretion. However, control of intragastric ph was similar for esomeprazole and omeprazole at a high dose of 80 mg (over 30 min) + 8 mg/h (for 23.5 h) [68]. Esomeprazole 40 mg i.v. resulted in 11.8 h with an intragastric ph>4 compared with 5.6 h (P<0.0001) and 7.2 h(p<0.001) for pantoprazole 40 mg i.v. as infusion or bolus injection, respectively, suggesting that esomeprazole was more potent than pantoprazole [69]. Likewise, oral administration of esomeprazole 40 mg b.i.d. provided better and more consistent intragastric acid control than pantoprazole 40 mg b.i.d. [70]. In patients with GERD, esomeprazole (40 mg daily or b.i.d.) was compared with lansoprazole (30 mg daily or b.i.d.). Mean time ph>4 and mean 24-h ph were highest for esomeprazole 40 mg b.i.d., followed by lansoprazole 30 mg b.i.d., esomeprazole 40 mg once daily, and lansoprazole 30 mg once daily [71]. Based on three pooled open studies in 80 subjects, single doses of rabeprazole (20 mg) and esomeprazole (40 mg) were equivalent in their effects on 24-h intragastric ph. After 5 days dosing, rabeprazole (20 mg) maintained ph>4 longer than 20 mg esomeprazole (62% vs 56%; P=0.046), whereas the lower dose (10 mg) of rabeprazole was slightly less effective (48%; P=0.035) [72]. These results were consistent with two other RCTs [73, 74]. On a milligram basis, acid inhibition by rabeprazole was significantly superior to omeprazole. After the initial dose, rabeprazole (20 mg) inhibited acid secretion by 72% after 11 h and by 64% after 23 h compared with omeprazole 20 mg (49% and 38%, P<0.03). This advantage was maintained for 8 days of treatment [75]. In addition, single and multiple doses of rabeprazole 20 mg produced greater acid suppression than oral or i.v. pantoprazole 40 mg [76, 77]. Furthermore, a reduced dose of rabeprazole (10 mg b.i. d. ) was comparable with the higher dosages of rabeprazole (20 mg b.i.d. ), to lansoprazole (30 mg b.i.d. ), and to omeprazole (20 mg b.i.d. ) for acid-suppressive efficacy [78, 79]. Overall, pharmacodynamic and clinical data indicated that on a milligram basis, rabeprazole can provide the greatest degree of acid suppression among the available PPIs. It also may provide a faster onset toward a maximal antisecretory effect than other drugs of this class. This can be of therapeutic relevance, as clinical outcome depends on extent and duration of secretory inhibition. Impact of CYP2C19 polymorphism on acid suppression PPI-induced inhibition of acid secretion is closely related to AUC values [3, 14]. As EMs of CYP2C19 have smaller AUC values than PMs, the genotype-dependent difference in pharmacokinetics will translate directly into a less pronounced acid suppression in EMs (about 70% of Caucasian patients) compared with PMs (see Table 2), so that EMs (and carriers of the CYP2C19*17 allele) will demonstrate a higher rate of therapeutic nonresponse. This topic has been extensively investigated [80, 81]. In eight healthy EMs, rabeprazole (10 mg/day) showed a faster onset of rising intragastric ph and a stronger inhibition of gastric acid secretion than did lansoprazole (30 mg/day) or omeprazole (20 mg/day) [82]. In nine healthy, H. pylori-negative homems treated with rabeprazole, omeprazole and lansoprazole once daily at reduced Table 2 Influence of CYP2C19 genotype on intragastric ph (according to Klotz [3]) Median ph over 24 h PPI PM hetem homem Omeprazole (20 mg for 7/8 days) Lansoprazole (30 mg for 8 days) Rabeprazole (20 mg for 8 days) PPI proton pump inhibitor, PM poor metabolizers, hetem heterozygous extensive metabolizers, homem homozygous extensive metabolizers

6 940 Eur J Clin Pharmacol (2008) 64: and standard doses for 7 days, the median values of the 24-h percent of time at ph>4 was dose dependent but did not exceed 65% under any of the regimens tested. Thus, to achieve effective acid suppression for the initial therapy of GERD, higher doses were needed in homems [83]. The efficacy of omeprazole (10 or 20 mg for 7 days) has been shown to be significantly stronger in PMs and hetems than in homems [84]. Likewise, in 31 Korean patients with GERD receiving omeprazole 20 mg daily for 28 days, 24-h intragastric ph in PMs (5.3) was higher (P<0.005) than that in homems (2.8) and hetems (3.6) [85]. In healthy volunteers (six homems, nine hetems, five PMs) treated with lansoprazole 30 mg b.i.d., the median of 24-h intragastric ph in PMs (6.1) was higher (P<0.05) than those in homems (4.5) and hetems (5.0). In contrast, when lansoprazole 30 mg b.i.d. was given with famotidine 20 mg b.i.d., the median intragastric ph was 5.4, 5.7, and 6.1, respectively. Thus, acid inhibition by lansoprazole was significantly influenced by the CYP2C19 genotype, but apparently this genetic influence could be offset by the concomitant use of the H 2 -receptor antagonist famotidine [86]. The impact of the CYP2C19 genotype on the efficacy of rabeprazole was less consistent. In H. pylori-negative GERD subjects given rabeprazole 10 mg/day for 8 weeks, the intragastric ph elevation was independent of CYP2C19 genotypes [87], which was consistent with another study [52]. In contrast, two studies with rabeprazole 20 or 40 mg daily for 8 days demonstrated that acid inhibition by rabeprazole was dependent on the CYP2C19 genotype [88, 89]. In Korean patients, pantoprazole 40 mg daily exhibited a variable acid inhibition that was significantly dependent on the CYP2C19 genotype [90]. As there is a significantly positive relationship between the extent and duration of elevated intragastric ph and the clinical efficacy of PPIs, it can be anticipated that the CYP2C19 genotype will have a significant impact on the therapeutic outcome of three PPIs (e.g., omeprazole, lansoprazole, pantoprazole) that are predominantly metabolized by this polymorphic enzyme. Therapeutic use Peptic ulcer For the management of PU (healing induction and remission maintenance), both suppression of gastric acid secretion and eradication of H. pylori are important mainstays. In general, there is little overall difference in PU healing rates among PPIs, and reported small differences can be explained best by the variable dosage regimens applied [91 93]. H. pylori eradication is accomplished efficiently (80 90% eradication rates) by standard triple therapy with a PPI, amoxicillin, and clarithromycin or metronidazole [94 96] (Table 3), and there was no significant difference among the PPIs [97 101]. A consensus on the optimal duration (7, 10, or 14 days) of first-line triple therapy is still missing. However, many controlled studies and meta-analyses [ ] indicated that extending this strategy beyond 7 days is very unlikely to be clinically useful, especially if taking antibioticinduced AEs into account [101]. There are even some data suggesting that a rabeprazole-based triple therapy for 4 days will result in eradication rates around 90% [109, 110]. More recently, sequential treatment consisting of 5 days of therapy with a PPI and amoxicillin followed by 5 days of PPI + clarithromycin and tinidazole has attracted some favorable attention [ ]. As already outlined, the genotype of CYP2C19 affects the pharmacokinetics and pharmacodynamics of most PPIs. In a recent meta-analysis of dual and triple therapies a, marked difference in the H. pylori eradication rates between PMs/hetEMs vs. homems was calculated, especially for omeprazole [115]. Therefore, CYP2C19 polymorphism is a major predictor of treatment failure for H. pylori eradication [116, 117]. Table 3 Recommendation of Helicobacter pylori eradication formulated in the Maastricht Consensus Report (adapted from Malfertheiner et al. [94]) Choice First-choice treatment Second-choice treatment Third-choice treatment (rescue treatment) Statements PPI-amoxicillin-clarithromycin or metronidazole treatment remains the recommended first-choice treatment in populations with less than 15 20% clarithromycin resistance prevalence. In populations with less than 40% metronidazole resistance prevalence PPI-clarithromycin-metronidazole is preferable. Quadruple therapies are alternative first-choice treatments. Bismuth-based quadruple therapies remain the best second-choice treatment. If not available, a PPI, amoxicillin, or tetracycline and metronidazole are recommended. Rescue treatment should be based on antimicrobial susceptibility testing.

7 Eur J Clin Pharmacol (2008) 64: Reflux disease The objective of treating GERD [118] is to relieve troublesome symptoms (e.g., heartburn, regurgitation), to restore quality of life, to heal oesophagitis (if present), and to reduce the risk of complications. Thereby, acid suppression is the mainstay of therapy, and PPIs represent the best choice because they are more potent than H 2 -receptor antagonists [ ]. Depending on the administered dose of the PPI and treatment duration, e.g., 4 or 8 weeks, healing in around 50 60% and 80 90% of patients, respectively, has been observed [ ]. Thereby, symptom relief was apparently somewhat faster with rabeprazole than with omeprazole [127, 128]. Healing of GERD can be restored either by maintenance (remission rates about 85%) or, alternatively, by on-demand treatment [ ]. Several studies have demonstrated the efficacy of pantoprazole (20 mg), esomeprazole (20 mg), or rabeprazole (10 mg) as on-demand therapy for long-term management of patients with mild GERD [133, 134]. The impact of CYP2C19 polymorphism on the clinical outcome was also evaluated in patients with GERD. The presence of a variant allele was associated with a significantly lower risk of gastric-acid breakthroughs during PPI therapy [135]. Likewise, in two independent trials, 8-week healing rates of lansoprazole (30 mg) were significantly higher in PMs (85% and 100%) and hetems (68% and 95%) compared with homems (46% and 77%) [136, 137]. Furthermore, during maintenance therapy (15 mg lansoprazole) remission rates (after 6 months) were 61.5%, 78%, and 100% in homems, hetems, and PMs, respectively [138]. In contrast, in three studies with esomeprazole, rabeprazole, and omeprazole, healing rates were not significantly affected by the CYP2C19 genotype [13, 139, 140]. In patients with nonerosive reflux disease (NERD) [118], the response to standard doses of PPIs seems to be for yet unknown reasons lower than in patients with GERD [141, 142]. Symptoms have been improved by rabeprazole (10 mg), esomeprazole (20 mg), and lansoprazole (15 mg), and on-demand treatment with PPIs appears to be also effective in the long-term management of NERD [129, 133, ]. NSAID-induced gastrointestinal lesions NSAIDs can cause GI lesions and result in dyspeptic symptoms and ulcerations and lead to increased risk of serious GI complications. Factors associated with an increased risk include intake of low-dose aspirin (ASS), history of ulcer or upper GI bleeding, age > 70 years, and concomitant use of NSAIDs [147, 148]. Therefore, patients at risk should be considered for alternatives to NSAID therapy, modifications of risk factors, as well as preventive strategies such as cotherapy with gastroprotective agents (PPIs or misoprostol) or cyclo-oxygenase-2 (COX-2)- selective inhibitors (coxibs) [149]. Importantly, prevention strategies must take into account both GI and cardiovascular (CV) risk factors, as coxibs and probably most traditional NSAIDs increase the incidence of serious CV events [150, 151]. Cotherapy with any PPI is an established option for prevention and healing of NSAID-associated GI lesions. All PPIs can provide effective acid suppression and decrease the morbidity and mortality associated with GI lesions [152, 153]. One meta-analysis of 118 trials (76,322 patients) assessed the relative effectiveness and cost-effectiveness of five strategies for the prevention of NSAID-induced GI toxicity. PPIs significantly reduced the risk of symptomatic ulcers [relative risk (RR) 0.09; 95% CI ] compared with placebo, and NSAIDs plus PPIs was the most cost-effective strategy for avoiding endoscopic ulcers in patients requiring long-term NSAIDs therapy [154]. PPIs will prevent back diffusion of hydrogen ions (the major aggressive factor) into the mucosa. In addition, as shown for rabeprazole, the mucosa might be restored by an increase in gastric production of protective mucus and mucin [155, 156]. Recent concern regarding potential CV risks of coxibs will favor PPI cotherapy with NSAIDs. The minor gastricsparing effect of coxibs is offset by concomitant use of lowdose ASS, whereas use of some NSAID plus PPI regimens may negate ASS s antiplatelet benefits. In addition, NSAIDs plus PPIs is at least as effective as the coxib strategy and may be more cost effective [157]. According to a meta-analysis of four studies, NSAIDs plus PPIs afforded greater risk reduction for dyspepsia than did coxibs, and NSAIDs plus PPIs vs. NSAIDs alone revealed a 66% RR reduction for PPI cotherapy with an absolute risk reduction of 9% [158]. Consequently, PPI should be considered for the treatment and prevention of NSAIDsinduced dyspepsia. It has been found that H. pylori infection and NSAID use represent independent and synergistic risk factors for PU [159, 160]. In a meta-analysis of 21 studies, uncomplicated PU was more common in H. pylori-positive than H. pylori-negative NSAID users [odds ratio (OR) 1.81]. In six age-matched controlled studies, H. pylori infection and NSAIDs use significantly increased the risk of PU (OR 4.03 and 3.10, respectively). The risk was 17.5-fold higher when both factors were present [160]. Evidence suggested that H. pylori eradication may prevent NSAID-induced ulcers in NSAID-naive patients, and in patients receiving long-term NSAIDs, PPIs were very effective in preventing ulcer recurrence [161, 162]. Peptic ulcer bleeding Peptic ulcer bleeding (PUB) represents an important emergency situation that is associated with considerable

8 942 Eur J Clin Pharmacol (2008) 64: morbidity, mortality, and healthcare system costs [163]. Several risk factors are involved, and elimination or modification of such etiopathogenetic factors will reduce the frequency of ulcer recurrences and PUB. Based on clinical and epidemiological data, it is evident that H. pylori infection and NSAID intake (including low-dose ASS and coxibs) are the two most important and independent risk factors [164, 165]. In addition, the elderly represent a population of elevated risks [164]. In three recent analyses, RR factors have been estimated (see Table 4). All NSAIDs and antiplatelet/anticoagulant agents have a similar potential to induce PUB, and if such drugs have to be taken, preventive strategies, including H. pylori eradication, should be applied [ ]. Several options for treating PUB and preventing rebleedings are available. Epinephrine injection is the most common endoscopic therapy for PUB [169], but following this successful management, rebleeding will occur in about 20% of patients [170]. Meta-analyses indicate that endoscopic hemostasis has reduced rebleeding and surgical interventions by >60% and mortality by 45% [171]. Because profound acid suppression (intragastric ph>6.0) optimizes stability of blood clots overlaying the ulcer and reduces the risk of rebleeding, endoscopic hemostasis is often combined with drug-induced (H 2 -receptor antagonists, PPIs) reduction of gastric acidity [172, 173]. Several reviews and meta-analyses of RCTs have been published concerning whether high-dose PPIs (intravenously as bolus/infusion and/or orally) affect PUB (see Table 5). Assessed outcomes were most frequently 30-days mortality, rebleeding, need for surgery, or endoscopic retreatment. PPIs significantly reduced the risks of rebleeding and the need for surgery. However, mortality was not affected by PPI treatment. Surprisingly, this hard outcome measure was reduced in Asian trials [ ]. Because most PPIs are primarily metabolized by the polymorphic CYP2C19, and because PMs are much most frequent (about 20%) in Asians than in Caucasians (about 3%), it could be speculated that in the Asian population, drug exposure by the high standard doses of PPI is more pronounced and thus more effective. Recently it was shown by a randomized prospective study in 129 bleeding peptic ulcer patients that oral treatment with rabeprazole (20 mg b.i.d. ) was equally effective as endoscopic hemoclipping with subsequent i.v. administration of H 2 -receptor antagonists in terms of hemostasis (93%) and rebleeding (7%) rates [177]. Besides PPIs, (RR: 0.33) H 2 -receptor antagonists (RR: 0.65) and nitrates (RR: 0.52) can also reduce the risks for PUB associated with NSAID intake, low-dose ASS, and clopidogrel [178]. There is some evidence that i.v. PPI therapy is more effective than oral treatment. However, the higher costs involved with i.v. administration do not justify this preference [179]. In conclusion, the addition of PPIs for endoscopic treatment of PUB is justified only for patients who have a high-risk lesion at endoscopy, as so far, there are no convincing data concerning the reduction of mortality. In addition, testing for and cure of H. pylori infection should be considered, as this represents an independent risk factor. Zollinger-Ellison syndrome The Zollinger-Ellison syndrome (ZES; gastrinoma) is characterized by hypersecretion of acid and intestinal ulcerations. Acid output and serum gastrin levels are elevated several fold in these patients. A proportion of subjects (up to 40%) may be cured by resection of the gastrinoma. In addition, gastric-acid hypersecretion and the syndrome can be effectively controlled by PPIs [180]. In general, high doses of PPIs (initially often administered intravenously by a short infusion) have been applied and Table 4 Relative risk (RR) factors for peptic ulcer bleeding Factor RR factor (95% CI) Analyzed cases Reference tnsaids a 3.7 ( ) 1,561 García Rodríguez & Barreales Tolosa [168] Coxibs a 2.6 ( ) NSAIDs 5.3 ( ) 2,777 Lanas et al. [166] Rofecoxib 2.1 ( ) Clopidogrel/ticlopidine 2.8 ( ) ASS (100 mg/day) 2.7 ( ) Anticoagulants 2.8 ( ) Preexisting peptic ulcer 4.3 (P=0.043) 822 Fisher et al. [167] Smoking 3.1 (P=0.023) Use of antiplatelet agents 6.5 (P=0.046) tnsaids/coxibs 4.9 (P=0.060) CI confidence interval, tnsaids traditional nonsteroidal anti-inflammatory drugs a Dose-dependent effects

9 Eur J Clin Pharmacol (2008) 64: Table 5 Proton pump inhibitor (PPI) therapy for peptic ulcer bleeding: summary of recent reviews and meta-analyses Assessed outcome (after 30 days) Odds ratio (95% CI) for effects of high-dose PPI a compared with placebo Number of trials (number of patient) b References Mortality (all trials) 1.01 ( ) 24 (4,373) Leontiadis et al. [ ] Mortality (7 Asian trials) 0.35 ( ) Rebleeding 0.49 ( ) Endoscopic retreatment 0.32 ( ) Mortality 1.12 ( ) 4 (1,512) Dorward et al. [208] Rebleeding 0.81 ( ) Need for surgery 0.96 ( ) Mortality (all causes) 1.02 ( ) 26 (4,670) Khuroo et al. [209] Ulcer deaths 0.58 ( ) Nonulcer deaths 1.60 ( ) Mortality 2.7% 18 (1,855) Bardou et al. [210] Rebleeding 14.6% Need for surgery 5.4% Mortality 1.11 ( ) 21 (2,915) Leontiadis et. [211, 212] Rebleeding 0.46 ( ) Need for surgery 0.59 ( ) Mortality and need for surgery Both rates were not different 3 (1,045) Andriulli et al. [213] Rebleeding 0.50 ( ) a Intravenous bolus (40 80 mg) and infusion (at least 6 8 mg/h) b It must be emphasized that the same trials (patients) have frequently been analyzed control of acid secretion (measurement of acid output; 24-h intragastric ph monitoring) was the primary efficacy endpoint. Doses were generally adjusted (titrated) according to the individual response in lowering basal acid output. In a short-term (7 10 days) open study with 11 male patients with ZES, omeprazole ( mg/day, mean dose 63 mg), lansoprazole ( mg/day, mean dose 75 mg), and pantoprazole ( mg/day, mean dose 116 mg) demonstrated a comparable antisecretory effect [181]. In 46 ZES patients during long-term (up to 10 years) treatment, the median effective lansoprazole dose to control acid secretion was approximately 80 mg/day (range mg/day) [182]. Likewise, 49 patients with ZES and eight hypersecretors were treated with a daily median dose of 75 mg lansoprazole ( mg/day) for up to 13 years. Forty-seven patients (70%) remained symptom and lesion free, and 90% of patients had good to excellent clinical outcomes without surgery [183]. In 26 patients with ZES and in nine with idiopathic hypersecretion, control of acid secretion was achieved by individual treatment with pantoprazole at 6 months, with doses of 40 mg b.i.d. (24), 80 mg b.i.d. (7), and 120 mg b.i.d. (2); it failed in two subjects [184]. This study was extended to an observation for 3 years (24). With maintenance doses between 40 and 120 mg b.i.d., acid output was controlled in all patients [185]. In 20 patients with ZES or idiopathic hypersecretion, acid output was also controlled by rabeprazole (60 mg for most patients) for 2 years. Gastric biopsies showed no enterochromaffin-like (ECL) cell dysplasia or neoplasia [186]. Form these limited clinical data, it can be concluded that high, individualized doses of PPIs offer a (long-term) treatment option for patients with ZES. Clinical safety and adverse effects PPIs have demonstrated an excellent safety profile after approximately 20 years of clinical use in millions of patients with acid-related diseases, including pregnant women and children [187, 188]. PPIs are generally well tolerated, and the incidence of AEs is relatively low. The most frequent AEs, with incidences of 1 3% reported in the clinical trials, were headache, nausea, diarrhea, skin rashes, and constipation [189]. Serious AEs are infrequent, although rare cases of toxic hepatitis and visual disturbance have been reported [189, 190]. An increased risk of community-acquired pneumonia associated with PPI treatment has been found in a large cohort of patients [191]. Interestingly, in a population-based case-control study, the initiation of treatment with PPIs showed a particularly strong association with communityacquired pneumonia (OR 5.0; 95% CI ), whereas the risk decreased with treatment (OR 1.3; 95% CI, ) [192]. Some reports indicated that acute interstitial nephritis (AIN) was associated with PPIs [193, 194]. However, a recent systematic review of 64 cases demonstrated that PPI-related AIN was rare, idiosyncratic, and difficult to predict; there was a low prevalence association [195]. In addition, whether PPIs are a risk factor for

10 944 Eur J Clin Pharmacol (2008) 64: Clostridium difficile-associated disease is controversial [196, 197]. Because long-term antisecretory therapy can affect absorption of calcium, PPIs may increase indirectly the risk of hip fracture, particularly for patients taking more than one dose per day [198, 199]. One nested case control study showed that the OR for hip fractures increased over time with PPI use: 1.22, 1.41, 1.54, and 1.59 for 1, 2, 3, and 4 years, respectively [199]. Of major interest is PPI-induced hypoacidity and hypergastrinemia. Long-term hypergastrinemia can cause ECL cell hyperplasia, which may predispose to carcinoids and might be important in gastric carcinogenesis [200]. In addition, PPI therapy affects the pattern and severity of H. pylori gastritis and accelerates the process of corpus gland loss, which is considered a risk factor for the development of gastric cancer [201]. To date, long-term PPI therapy was not associated with an increased risk of gastric or colorectal cancer [ ], but apparently it was associated with an up to fourfold increase in the risk of fundic gland polyps [204]. Frequency and size of adenomatous polyps in the colon cases (116) that used PPIs were not different from matched controls (194) in a series of 2,868 consecutive patients undergoing colonoscopies [205]. Conclusions Since the introduction of PPIs, considerable progress has been achieved in the management of gastric-acid-related disorders (e.g., PU, GERD, ZES) and NSAID-induced GI lesions. The success of PPIs is due both to their welldocumented clinical efficacy and safety. As the active principles of available PPI-prodrugs have the identical mode of action, resulting in elevating intragastric ph, the various PPIs can be used in an interchangeable fashion. Whereas the pharmacodynamic profile of PPIs is the same, some pharmacokinetic differences exist among these valuable agents that can slightly modify their clinical action because there are close relationships between pharmacokinetics, pharmacodynamics, and therapeutic outcome. Systemic drug exposure (AUC) will determine how fast and how long the intragastric target ph can be maintained above the threshold (target) level of >3 (PU) or >4 (GERD), which subsequently affects the healing rates of acid-related disorders, including H. pylori infections and NSAID-induced GI lesions. Similar to other drugs, PPIs demonstrate a large interindividual variability in drug disposition (e.g., bioavailability, AUC, metabolic pathways, hepatic elimination rates), which is caused by genetic and nongenetic factors. In addition, the interaction potential of PPIs in terms of drug metabolism shows some compound-specific differences. In relation to genetic factors, the influence of polymorphisms of CYP2C19 on the clinical outcome of PPIs has been extensively investigated. As this enzyme is mainly responsible for the metabolism (hepatic elimination rate) of omeprazole, lansoprazole, and pantoprazole, their AUCs and pharmacodynamic responses are much more dependent on the genotype/phenotype of CYP2C19 compared with esomeprazole (metabolic contribution by CYP3A4) or rabeprazole (eliminated mainly by a nonenzymatic pathway). It has been well documented by several clinical studies that standard doses of omeprazole and lansoprazole are too low for many EMs (especially carriers of the CYP2C19*17 mutation), resulting in a larger proportion of nonresponders in this subgroup representing about 70% of Caucasian populations. Thus, genotype-adjusted dosing could improve the clinical outcome of these PPIs. Likewise, a PPI should be preferred that is less dependent on CYP2C19 to overcome this genetically controlled variability [206]. As all PPIs dose (AUC)-dependently elevate intragastric ph, solubility and absorption of other drugs given concomitantly might be affected to some extent, as has been exemplified for ketoconazole, vitamin B 12,digoxin,atazanavir, or calcium. In contrast to this class effect, drug interactions on the level of hepatic metabolism are more compound specific. With omeprazole, induction of the CYP1A subfamily has been reported, and inhibition of hepatic elimination (mainly via CYP2C19 and CYP3A4) of several drugs must be considered (the latter is seen to some extent with esomeprazole also). The other three PPIs are apparently devoid of this metabolic interaction potential. As the liver represents the major site of elimination for all PPIs, patients with moderate or severe hepatic dysfunction have a reduced CL, and dosage might be reduced for pharmacokinetic reasons by factor 2 in such subjects. Due to structural differences affecting the biophysical and biochemical properties of the PPIs, different dosages of the various agents are needed for the same pharmacodynamic response, e.g., for control of acid secretion in patients with ZES, relative potency dose ratios can be calculated for omeprazole, lansoprazole, and pantoprazole (1:0.81:0.54). In many cases, PPIs have to be taken long term. Under such conditions, hypergastrinemia and ECL cell hyperplasia can develop. However, so far, this has not been associated with a significantly increased risk of gastric cancer. As experience for almost 20 years is now available, this potential safety concern regarding all effective antisecretory strategies and conditions should not limit the use of PPIs. For various reasons (e.g., nonadherence, CYP2C19*1 or *17 EM-genotype, resistance), some patients will be refractory to PPI treatment [207]. Such nonresponders might profit from an increased dose, a b.i.d. regimen, the addition of an H 2 -receptor antagonist, or switching to a PPI with higher potency and less impact of genetic poly-

11 Eur J Clin Pharmacol (2008) 64: morphisms. In addition, developing novel PPIs or other antisecretory agents might be an option for the future. In conclusion, based on long-term experience and extensive clinical data, PPIs still represent drugs of first choice for managing PU-including H. pylori infection, GERD, ZES, and NSAID-induced GI lesions. They provide a safe and cost-effective treatment. Acknowledgements The secretarial help of Mrs. U. Hengemühle is appreciated. This work was supported by the Robert Bosch Foundation, Stuttgart, Germany, and an educational grant from Eisai Europe Ltd., London. Conflict of Interest of interest. References The authors declare that they have no conflicts 1. Sachs G, Shin JM, Howden CW (2006) Review article: the clinical pharmacology of proton pump inhibitors. Aliment Pharmacol Ther 23(Suppl 2): Furuta T, Shirai N, Sugimoto M, Ohashi K, Ishizaki T (2004) Pharmacogenomics of proton pump inhibitors. Pharmacogenomics 5: Klotz U (2006) Clinical impact of CYP2C19 polymorphism on the action of proton pump inhibitors: a review of a special problem. Int J Clin Pharmacol Ther 44: Robinson M (2005) Proton pump inhibitors: update on their role in acid-related gastrointestinal diseases. Int J Clin Pract 59: Lassen AT (2007) Acid-related disorders and use of antisecretory medication. Dan Med Bull 54: Klotz U (2000) Pharmacokinetic considerations in the eradication of Helicobacter pylori. Clin Pharmacokinet 38: Esplugues JV, Martí-Cabrera M, Ponce J (2006) Safety of proton pump inhibitors. Med Clin (Barc) 127: Maffei M, Desmeules J, Cereda JM, Hadengue A (2007) Side effects of proton pump inhibitors (PPIs). Rev Med Suisse 123: , Horn JR, Howden CW (2005) Review article: similarities and differences among delayed-release proton-pump inhibitor formulations. Aliment Pharmacol Ther 22(Suppl 3): Sostek MB, Chen Y, Andersson T (2007) Effect of timing of dosing in relation to food intake on the pharmacokinetics of esomeprazole. Br J Clin Pharmacol 64: Hassan-Alin M, Andersson T, Niazi M, Röhss K (2005) A pharmacokinetic study comparing single and repeated oral doses of 20 mg and 40 mg omeprazole and its two optical isomers, S- omeprazole (esomeprazole) and R-omeprazole, in healthy subjects. Eur J Clin Pharmacol 60: Simon B, Müller P, Pascu O, Gatz G, Sander P, Huber R, Mascher H (2003) Intra-oesophageal ph profiles and pharmacokinetics of pantoprazole and esomeprazole: a crossover study in patients with gastro-oesophageal reflux disease. Eur J Gastroenterol Hepatol 15: Schwab M, Klotz U, Hofmann U, Schaeffeler E, Leodolter A, Malfertheiner P, Treiber G (2005) Esomeprazole-induced healing of gastroesophageal reflux disease is unrelated to the genotype of CYP2C19: evidence from clinical and pharmacokinetic data. Clin Pharmacol Ther 78: Klotz U, Schwab M, Treiber G (2004) CYP2C19 polymorphism and proton pump inhibitors. Basic Clin Pharmacol Toxicol 95: Xie HG, Stein CM, Kim RB, Wilkinson GR, Flockhart DA, Wood AJ (1999) Allelic, genotypic and phenotypic distributions of S-mephenytoin 4 -hydroxylase (CYP2C19) in healthy Caucasian populations of European descent throughout the world. Pharmacogenetics 9: Desta Z, Zhao X, Shin JG, Flockhart DA (2002) Clinical significance of the cytochrome P450 2C19 genetic polymorphism. Clin Pharmacokinet 41: Uno T, Niioka T, Hayakari M, Yasui-Furukori N, Sugawara K, Tateishi T (2007) Absolute bioavailability and metabolism of omeprazole in relation to CYP2C19 genotypes following single intravenous and oral administrations. Eur J Clin Pharmacol 63: Qiao HL, Hu YR, Tian X, Jia LJ, Gao N, Zhang LR, Guo YZ (2006) Pharmacokinetics of three proton pump inhibitors in Chinese subjects in relation to the CYP2C19 genotype. Eur J Clin Pharmacol 62: Hu XP, Xu JM, Hu YM, Mei Q, Xu XH (2007) Effects of CYP2C19 genetic polymorphism on the pharmacokinetics and pharmacodynamics of omeprazole in Chinese people. J Clin Pharm Ther 32: Ishizawa Y, Yasui-Furukori N, Takahata T, Sasaki M, Tateishi T (2005) The effect of aging on the relationship between the cytochrome P450 2C19 genotype and omeprazole pharmacokinetics. Clin Pharmacokinet 44: Sim SC, Risinger C, Dahl ML, Aklillu E, Christensen M, Bertilsson L, Ingelman-Sundberg M (2006) A common novel CYP2C19 gene variant causes ultrarapid drug metabolism relevant for the drug response to proton pump inhibitors and antidepressants. Clin Pharmacol Ther 79: Baldwin RM, Ohlsson S, Pedersen RS, Mwinyi J, Ingelman- Sundberg M, Eliasson E, Bertilsson L (2008) Increased omeprazole metabolism in carriers of the CYP2C19*17 allele; a pharmacokinetic study in healthy volunteers. Br J Clin Pharmacol 65: Hunfeld NG, Mathot RA, Touw DJ, van Schaik RH, Mulder PG, Franck PF, Kuipers EJ, Geuss WP (2008) Effect of CYP219*2 and *17 mutations on pharmacodynamics and kinetics of proton pump inhibitors in Caucasians. Br J Clin Pharmacol 65: Marier JF, Dubuc MC, Drouin E, Alvarez F, Ducharme MP, Brazier JL (2004) Pharmacokinetics of omeprazole in healthy adults and in children with gastroesophageal reflux disease. Ther Drug Monit 26: Zhao J, Li J, Hamer-Maansson JE, Andersson T, Fulmer R, Illueca M, Lundborg P (2006) Pharmacokinetic properties of esomeprazole in children aged 1 to 11 years with symptoms of gastroesophageal reflux disease: a randomized, open-label study. Clin Ther 28: Li J, Zhao J, Hamer-Maansson JE, Andersson T, Fulmer R, Illueca M, Lundborg P (2006) Pharmacokinetic properties of esomeprazole in adolescent patients aged 12 to 17 years with symptoms of gastroesophageal reflux disease: A randomized, open-label study. Clin Ther 28: Blume H, Donath F, Warnke A, Schug BS (2006) Pharmacokinetic drug interaction profiles of proton pump inhibitors. Drug Saf 29: Arayne MS, Sultana N, Qureshi MS, Naseem S (2006) Interaction studies of omeprazole with mefloquine, pyrimethamine and sulfadoxine. Pak J Pharm Sci 19: Kang BC, Yang CQ, Cho HK, Suh OK, Shin WG (2002) Influence of fluconazole on the pharmacokinetics of omeprazole in healthy volunteers. Biopharm Drug Dispos 23: Yasui-Furukori N, Takahata T, Nakagami T, Yoshiya G, Inoue Y, Kaneko S, Tateishi T (2004) Different inhibitory effect of fluvoxamine on omeprazole metabolism between CYP2C19 genotypes. Br J Clin Pharmacol 57: