Professor J. Horn, Department of Pharmacy, University of Washington, Box 357630, Seattle, WA 98195–7630, USA. E-mail: firstname.lastname@example.org
Proton pump inhibitors are now considered the mainstay of treatment for acid-related disease. Although all proton pump inhibitors are highly effective, the antisecretory effects of different drugs in this class are not completely consistent across patients. One reason for this is the acid-suppressing effect of Helicobacter pylori infection, which may augment the actions of proton pump inhibitors. A second important reason for interpatient variability of the effects of proton pump inhibitors on acid secretion involves genetically determined differences in the metabolism of these drugs. This article focuses on the impact of genetic polymorphism of cytochrome P450 (CYP)2C19 on the pharmacokinetics and pharmacodynamics of proton pump inhibitors, particularly rabeprazole. Results reviewed indicate that the metabolism and pharmacokinetics of rabeprazole differ significantly from those of other proton pump inhibitors. Most importantly, the clearance of rabeprazole is largely nonenzymatic and less dependent on CYP2C19 than other drugs in its class. This results in greater consistency of pharmacokinetics for rabeprazole across a wide range of patients with acid-related disease, particularly those with different CYP2C19 genotypes. The pharmacodynamic profile for rabeprazole is also characterized by more rapid suppression of gastric acid secretion than with other proton pump inhibitors, which is also independent of CYP2C19 genotype. The favourable pharmacokinetic/pharmacodynamic profile for rabeprazole has been shown to result in high eradication rates for H. pylori in both normal and poor metabolizers. Pharmacodynamic results have also suggested that rabeprazole may be better suited than omeprazole as on-demand therapy for symptomatic gastro-oesophageal reflux disease. Finally, the use of rabeprazole is not complicated by clinically significant drug–drug interactions of the type that have been reported for omeprazole.
Proton pump inhibitors are now considered the mainstay of treatment for acid-related disease. Clinical trial results have clearly established the superior efficacy of these agents over histamine2-receptor antagonists in the treatment of patients with duodenal and gastric ulcers, and both erosive and nonerosive gastro-oesophageal reflux disease (GERD).1
The antisecretory and clinical effects of proton pump inhibitors are not completely consistent across patients, and there are several reasons for this. First, the presence of the highly prevalent Gram-negative gastric pathogen Helicobacter pylori has been shown to enhance the effect of proton pump inhibitors in lowering gastric pH. Eradication of H. pylori has led to increased 24-h gastric acidity, particularly nocturnal acidity, and decreased plasma gastrin.2 The mechanism(s) by which H. pylori influences acid secretion is not completely understood, but it has been suggested that H. pylori-associated inflammation of the gastric corpus tends to diminish acid secretion due to bacterial secretion and cytokines (e.g. interleukin-1) that inhibit parietal cells.3 It has also been reported that H. pylori infection suppresses expression of the genes encoding H+K+-adenosine triphosphatase (ATPase; the proton pump) in parietal cells.4H. pylori infection may have variable effects on intragastric pH depending on the body's immunological response; there is a reduced acidity in individuals who demonstrated a specific immunological response to infection characterized by elevated levels of IgA, IgG and IgM antibodies (pH 6 vs. pH 2 when no immunoglobulin was detected).5
The second major reason for the variable effects of proton pump inhibitors in reducing gastric acidity is interindividual differences in the pharmacokinetics of these drugs, which have been related to specific genetic polymorphisms that play key roles in their metabolism. Achievement of effective concentration of a drug at its site of action is strongly determined by pharmacokinetic processes (absorption, distribution, metabolism and elimination). A large number of genetic polymorphisms have been identified for drug-metabolizing enzymes that can affect drug concentrations and, thus, pharmacodynamic and clinical responses.6 Cytochrome P450 (CYP)2C19 mediates the major metabolic transformations for most proton pump inhibitors, and its genetic polymorphism can lead to significant phenotypic variation that can significantly influence the metabolism and pharmacodynamic actions of proton pump inhibitors by this particular isoenzyme of the P450 system.7 This article will review the metabolism of proton pump inhibitors, the degree to which genetic polymorphisms influence their pharmacokinetic profiles and how this variability influences response to treatment. Rabeprazole, a newer proton pump inhibitor, is the focus of this review, but results for other drugs in this class will also be outlined.
Pharmacokinetics and metabolism of proton pump inhibitors
Proton pump inhibitors are relatively rapidly absorbed after oral administration and reach peak plasma concentrations within a few hours of dosing. These drugs are also extensively absorbed and have high bioavailabilities, exceeding 50% in most cases (Table 1).8–10 There are, however, significant differences among proton pump inhibitors with respect to their hepatic metabolism and the potential impact of genetic polymorphisms.
Table 1. Pharmacokinetic profiles for proton pump inhibitors8
Pantoprazole 40 mg
Omeprazole 20 mg
Lansoprazole 30 mg
Rabeprazole 20 mg
Esomeprazole 40 mg
AUC, area under the plasma concentration–time curve; Cmax, maximum plasma concentration; t½, elimination half-life.
Yes (for oral doses of 15–60 mg)
Yes (for oral doses of 10–40 mg)
Factors that affect absorption
Food not studied
Hepatic metabolism of proton pump inhibitors
The elimination of all proton pump inhibitors involves hepatic oxidation mediated by the CYP450 enzymatic system, primarily CYP2C19.10,11 The clearance of rabeprazole also involves CYP2C19, but to a smaller extent: it undergoes an almost complete, mainly nonenzymatic metabolism with renal elimination of its metabolites. Both CYP3A4 and CYP2C19 contribute to the small fraction of rabeprazole's elimination that involves hepatic biotransformation.12,13 This unique aspect of the metabolism of rabeprazole as compared with other proton pump inhibitors means that it may be less susceptible to the influence of genetic polymorphisms for either CYP3A4 or CYP2C19.
Clinical relevance of CYP2C19 genetic polymorphism
In addition to the wild type (CYP2C19*1), eight variant alleles (CYP2C19*2 to CYP2C19*8) have been identified that can alter the activity of CYP2C19 (Figure 1).14,15 Individuals with different CYP2C19 genotypes are divided into two main phenotypes: extensive and poor metabolizers. It has been estimated that 3–5% of whites and African Americans and 12–100% of various Asian groups have reduced activity of CYP2C19; they are referred to as poor metabolizers.14
The genetic polymorphism for CYP2C19 has significant metabolic and clinical consequences. This polymorphism affects metabolism of the anticonvulsant mephenytoin, diazepam and certain antidepressants,14 and its effects on the pharmacokinetics of these drugs has been cited as a potential reason for treatment failure, toxicities and unexpected drug interactions (see below).
Impact of CYP2C19 polymorphism on the pharmacokinetics of rabeprazole
Results from pharmacokinetic studies have indicated only modest effects of the CYP2C19 genotype on the pharmacokinetics and metabolism of rabeprazole, contrasting sharply with findings for other proton pump inhibitors. The impact of CYP2C19 polymorphism on the pharmacokinetics of omeprazole, lansoprazole and rabeprazole has been assessed in an open, randomized, crossover study that included 18 Japanese subjects (six heterozygous and six homozygous extensive metabolizers, and six poor metabolizers for CYP2C19). Results for rabeprazole (Table 2) indicated no significant effect of CYP2C19 genotype on the area under the plasma–concentration time curve (AUC), maximum plasma concentration (Cmax) or elimination half-life (t½). In contrast, the AUC and Cmax for both lansoprazole and omeprazole increased in poor vs. extensive metabolizers.16
Table 2. Pharmacokinetic parameters for rabeprazole in normal and poor metabolizers. Adapted from Sakai et al.16
AUC, area under the plasma concentration–time curve; Cmax, maximum plasma concentration; HetEM, heterozygous extensive metabolizers; HomEM, homozygous extensive metabolizers; N.S., not significant; N.T., not tested because P > 0.05 by one-way analysis of variance; PM, poor metabolizers; t½, elimination half-life.
Each value represents the mean
1048.0 ± 210.2
1114 ± 74.6
1240.5 ± 182.6
449.0 ± 121.8
492.6 ± 67.9
388.3 ± 56.9
1.13 ± 0.14
1.39 ± 0.20
1.54 ± 0.28
618.3 ± 141.9
1061.8 ± 269.2
4587.1 ± 681.6
251.1 ± 46.2
623.1 ± 149.1
1070.2 ± 185.3
1.09 ± 0.08
1.18 ± 0.20
2.41 ± 0.15
2549.3 ± 371.9
3484.2 ± 567.3
9379.7 ± 978.2
849.3 ± 146.7
955.4 ± 197.3
1550.1 ± 218.8
2.01 ± 0.62
2.47 ± 0.29
3.77 ± 0.31
These results are consistent with earlier findings that compared the pharmacokinetics of rabeprazole and omeprazole (each administered at a dose of 20 mg/day for 7 days) in 15 Japanese male volunteers (nine extensive metabolizers and six poor metabolizers). After the last dose (day 7), the AUCs for omeprazole and rabeprazole were increased by 4.4- and 1.9-fold, respectively, in the poor metabolizers.17 The modest increase in the AUC for rabeprazole observed in this study is also consistent with more recent results indicating that exposure to this proton pump inhibitor is not substantially increased in poor metabolizers of CYP2C19. In six homozygous and six heterozygous extensive metabolizers and six poor metabolizers, the AUC and Cmax for rabeprazole significantly increased, and clearance significantly decreased in the poor metabolizers; however, these changes were much smaller than those noted for other proton pump inhibitors.18
All of these findings correspond well with results from an in vitro study that used human liver microsomes to assess the effects of CYP2C19 on the metabolism of rabeprazole, lansoprazole and omeprazole. Results showed that after incubation with these microsomes, the residual concentrations of omeprazole and lansoprazole, but not rabeprazole, correlated well with the CYP2C19 activities (r = 0.938, P < 0.005; r = 0.962, P < 0.001; and r = 0.661, P = 0.11, respectively).19
Results from one study have indicated a substantially greater impact of CYP2C19 polymorphism on the pharmacokinetics of rabeprazole than did the trials summarized in the preceding paragraphs. This trial compared the pharmacokinetics of rabeprazole (20 mg twice daily) in extensive and poor metabolizers who also had documented H. pylori infections. The mean AUC for rabeprazole on day 1 was significantly higher in the poor metabolizers than in extensive metabolizers (5357 ± 883 ng/h/mL vs. 1131 ± 512 ng/h/mL).20 Similar results were observed on day 4, with no significant difference noted between day 1 and day 4 of dosing. The only difference on day 4 compared with day 1 was the significantly higher plasma gastrin levels in the poor metabolizers (P < 0.05). It is possible that the substantial difference between these results and those obtained in other pharmacokinetic studies of rabeprazole in extensive and poor metabolizers may be due to the very high (40 mg/day) rabeprazole doses used in this trial.20
In summary, pharmacokinetic evaluation of rabeprazole in poor metabolizers indicates modest increases in exposure to this proton pump inhibitor. Comparative studies have consistently shown that the CYP2C19 genetic polymorphism has significantly less impact on plasma levels of rabeprazole than on other proton pump inhibitors. This consistency in the pharmacokinetics of rabeprazole across genetically diverse patients has the potential to reduce interpatient variability in both pharmacological and clinical effects, and it may also reduce the risk for toxicities associated with unexpectedly high drug concentrations in poor metabolizers.
Impact of CYP2C19 polymorphism on the pharmacodynamics of rabeprazole
The pharmacodynamic effects of proton pump inhibitors are closely related to the plasma levels of these drugs.21 Therefore, it would be expected that patient characteristics that increase drug concentrations might alter efficacy in suppressing gastric acid secretion. The limited impact of the CYP2C19 polymorphism on the pharmacokinetics of rabeprazole further suggests that the acid-suppressing effects of this proton pump inhibitor should be more consistent across patients than other drugs in this class.
The suppression of gastric acid secretion by omeprazole (20 mg/day), lansoprazole (30 mg/day) and rabeprazole (10 mg/day) was investigated in healthy, H. pylori-negative, homozygous and heterozygous extensive metabolizers.22 Intragastric pH monitoring indicated that the mean ratios of the 24-h pH ≥ 3 holding time was significantly longer compared with baseline for rabeprazole and lansoprazole starting at day 1 and for omeprazole starting at day 2 (Figure 2). Furthermore, on day 1 the inhibition of gastric acid secretion was superior for rabeprazole compared with lansoprazole. All these results suggest that in H. pylori-negative extensive metabolizers, 10-mg rabeprazole has a faster onset of action and provides more profound suppression of acid secretion than either lansoprazole or omeprazole.22
The pharmacodynamic effects of omeprazole and rabeprazole (each administered at a dose of 20 mg/day for 8 days) were compared in 15 healthy volunteers. After both single and multiple doses, intragastric pH was significantly dependent on the CYP2C19 genotype for omeprazole, but not for rabeprazole (Table 3).23 A comparison of rabeprazole and lansoprazole provided similar results. Twenty H. pylori-negative male volunteers with CYP2C19 genotype, who were determined to be homozygous or heterozygous extensive metabolizers or poor metabolizers, were treated with either rabeprazole (20 mg/day) or lansoprazole (30 mg/day) for 7 days. The median intragastric pH during proton pump inhibitor administration was significantly influenced by CYP2C19 genotype in patients treated with lansoprazole, but not in those treated with rabeprazole (Figure 3). The median pH and percentage of time for pH ≥ 4 during treatment with lansoprazole were significantly higher in the poor metabolizers than in either group of extensive metabolizers. This may confer an advantage in poor metabolizers; however, this category represents only 13–23% of the East Asian population and less than 6% of whites and African Americans.24
Table 3. Intragastric pH values after single and repeated doses of omeprazole (OME) and rabeprazole (RAB)23
The lack of effect of CYP2C19 genotype and phenotype variations on the pharmacodynamics of rabeprazole observed in healthy volunteers also extends to patients with acid-related disease. The acid-suppressing effects of rabeprazole (20 mg/day) and omeprazole (40 mg/day) were compared in 30 patients with duodenal ulcers, and CYP2C19 genotype was determined by polymerase chain reaction–restriction enzyme analysis. Mean intragastric pH measured over 24 h was not significantly influenced by CYP2C19 genotype in the patients who received rabeprazole, but it was significantly reduced in homozygous extensive metabolizers who received omeprazole.25
In summary, results from studies that have evaluated the influence of CYP2C19 genotype on the pharmacodynamic actions of rabeprazole and other proton pump inhibitors are consistent with the pharmacokinetic data summarized in the preceding section: CYP2C19 genotype and phenotype have significantly less impact on the acid-suppressing effects of rabeprazole than on other proton pump inhibitors for which metabolism is more dependent on this enzyme.
Clinical efficacy of rabeprazole in patients with different CYP2C19 genotypes and phenotypes
Only a small number of studies have evaluated the impact of the CYP2C19 genetic polymorphism on the clinical efficacy of proton pump inhibitors. However, its impact on the effectiveness of combination regimens including rabeprazole or other proton pump inhibitors with antibiotics for eradication of H. pylori has been assessed in a number of studies.
Eradication of H. pylori
Eradication of H. pylori is usually accomplished with a combination of a proton pump inhibitor and two or more antibiotics. The antibiotics most often used for the treatment of H. pylori infection are more stable at a higher pH, and coadministration of a proton pump inhibitor to rapidly increase pH to > 4 reduces the degradation of antibiotics such as amoxicillin and clarithromycin and increases their effectiveness in eradicating H. pylori.26 Rabeprazole has a very rapid onset of antisecretory action, and its effectiveness in combination therapy for H. pylori eradication is not significantly influenced by the CYP2C19 genetic polymorphism.
Rabeprazole also has a more rapid onset of antisecretory action than other drugs in its class. In vitro, rabeprazole inhibited H+K+-ATPase in isolated hog gastric vesicles more rapidly than did pantoprazole, lansoprazole or omeprazole.27In vivo, the onset of acid-suppressing action with rabeprazole in healthyH. pylori-negative volunteers was more rapid than with lansoprazole, pantoprazole and omeprazole, as reflected by attainment of a significantly higher gastric pH after the first dose (P ≤ 0.04 vs. other proton pump inhibitors).28 Furthermore, the ability of the first dose of rabeprazole to raise intragastric pH is not significantly influenced by CYP2C19 genotype.29 All of these results suggest that rabeprazole should be highly suited for combination therapy aimed at H. pylori eradication in patients with acid-related disease.
Rabeprazole was evaluated in 128 patients with H. pylori-positive gastritis or peptic ulcer with CYP2C19 genotype determined by the polymerase chain reaction–restriction fragment length polymorphism method. Patients were treated with a 7-day regimen that included rabeprazole 20 mg/day, amoxicillin 2000 mg/day and either clarithromycin 500 mg/day or metronidazole 400 mg/day. Eradication rates for patients in the clarithromycin and metronidazole treatment arms were 98.1% and 91.3%, respectively. The rates of eradication were not significantly influenced by CYP2C19 genotype.30 In 102 H. pylori-positive patients with gastric ulcers, a 7-day triple-therapy regimen with rabeprazole (10, 20 or 40 mg/day) plus amoxicillin (1500 mg/day) and clarithromycin (1000 mg/day) provided similar H. pylori eradication regardless of the rabeprazole dosage used. The CYP2C19 phenotype had no influence on the eradication rates in these patients (86 vs. 77% for extensive vs. poor metabolizers, respectively).31 Similar results were obtained in 116 patients with gastric or duodenal ulcers who received the aforementioned combination of rabeprazole, amoxicillin and clarithromycin. Eradication rates ranged from 87% in heterozygous extensive metabolizers to 80% in poor metabolizers, with no significant differences among groups.32
In patients who had failed to achieve H. pylori eradication with the combination of lansoprazole, amoxicillin and clarithromycin, 1 week of therapy with rabeprazole (20 mg/day) plus amoxicillin (1500 mg/day) and metronidazole (500 mg/day) was significantly better than 2 weeks with the combination of rabeprazole (40 mg/day) and amoxicillin (2000 mg/day) (82% vs. 59%, intent-to-treat analysis; 88% vs. 66%, per protocol analysis; P < 0.01 for both).33 The bacterial eradication with rabeprazole-containing regimens was not significantly influenced by either CYP2C19 genotype or antibiotic resistance.33 The lack of effect of CYP2C19 genetic polymorphism on rabeprazole effectiveness on H. pylori eradication rates was further demonstrated in a comparative study between rabeprazole and lansoprazole;34 187 H. pylori-positive patients with peptic ulcers received either rabeprazole (20 mg/day) or lansoprazole (60 mg/day) combined with amoxicillin (1500 mg/day) and clarithromycin (800 mg/day) for 1 week. The overall H. pylori eradication rates were similar for both treatments; however, when extensive and poor metabolizers were differentiated, the extensive metabolizer group tended to have a superior rate of eradication with rabeprazole compared with lansoprazole (89% vs. 78%; P = 0.079).34
In summary, the results of noncomparative and comparative trials demonstrate that rabeprazole is highly and consistently effective in triple combination with antibiotics for eradication of H. pylori, due in part to the rapid onset of action of rabeprazole compared with other proton pump inhibitors and the corresponding protection of antibiotics from low gastric pH, making them more stable. The consistent effectiveness of rabeprazole across CYP2C19 genotypes is attributable to the fact that, unlike other proton pump inhibitors, the metabolism of rabeprazole is largely independent of this enzyme.
Clinical efficacy in acid-related disease
Limited information is available about the impact of the CYP2C19 polymorphism on symptom relief or healing of erosions or ulcers in patients treated with proton pump inhibitors. Results from a study of 88 Japanese patients with erosive GERD treated with lansoprazole (30 mg/day) for 8 weeks indicated a significant impact of CYP2C19 genotype on ulcer healing. The healing rates were 57.1%, 69.2% and 72.7% at 4 weeks and 77.4%, 95.0% and 100% at 8 weeks in homozygous extensive metabolizers, heterozygous extensive metabolizers and poor metabolizers, respectively. At 8 weeks, the healing rate of erosive reflux oesophagitis was significantly lower in homozygous extensive metabolizers than in the two other groups (P < 0.05).35
In a second study of 65 Japanese patients with endoscopically proven GERD, lansoprazole 30 mg daily was administered for 8 weeks.36 Oesophagitis healing rates were 45.8%, 67.9% and 84.6% in the homozygous extensive metabolizers, heterozygous extensive metabolizers and poor metabolizers, respectively. The healing rate was significantly lower in homozygous extensive metabolizers than in the poor metabolizers (P < 0.04). Plasma lansoprazole concentrations were significantly different (P < 0.04) between the groups. The mean lansoprazole concentrations obtained 3 h after dosing were 312.3 ng/mL, 439.9 ng/mL and 745.4 ng/mL in the homozygous extensive metabolizers, heterozygous extensive metabolizers and poor metabolizers, respectively.
No clinical trials have evaluated the impact of CYP2C19 polymorphism on the symptom relief or healing rates of ulcers or erosions in patients treated with rabeprazole. However, comparison of the first-dose acid-suppressing effects of rabeprazole 20 mg vs. omeprazole 20 mg indicated that rabeprazole maintained intragastric pH > 3 and > 4 for longer periods than did omeprazole and resulted in a higher average pH during the first 6 h after treatment in both homozygous and heterozygous extensive metabolizers. These findings prompt the conclusion that rabeprazole may be more suitable for rapid relief of GERD symptoms than is omeprazole for on-demand therapy in patients with this disease.29
Proton pump inhibitors are very safe drugs that have been used in millions of patients with acid-related disease. However, these agents differ markedly in their risk for pharmacokinetic interactions that may result in significant toxicities. The risk for such interactions is closely related to the manner in which these drugs are metabolized in the liver and cleared from the body. Adverse drug interactions can result from induction (loss of therapeutic benefit) or inhibition (increased toxicity from excessive effect) of drug elimination.37
Retrospective analysis of data for omeprazole and lansoprazole indicated that 58% of patients who were taking one of these proton pump inhibitors used at least one other drug with the potential for interaction. Risk for adverse events resulting from drug interactions were highest for patients taking a proton pump inhibitor and warfarin, clarithromycin, corticosteroids, carbamazepine, nifedipine or diclofenac.38 It has been noted further that poor metabolizers of proton pump inhibitors who lack CYP2C19 may be at particularly high risk for drug–drug interactions involving proton pump inhibitors.39 These patients will accumulate higher concentrations of the proton pump inhibitor, resulting in a greater effect of the proton pump inhibitor on other drugs.
Rabeprazole has a very low risk for pharmacokinetic interactions. It has been studied extensively and has been shown to be free of pharmacokinetic interactions that might result from CYP induction or inhibition.11,40 Results for rabeprazole indicate no significant interactions with theophylline, phenytoin, warfarin or diazepam. Rabeprazole does, like all other proton pump inhibitors, interact with ketoconazole (decreased absorption) and digoxin (minor degree of increased absorption).11
The results reviewed in this article indicate that the metabolism and pharmacokinetics of rabeprazole differ significantly from other proton pump inhibitors. Most importantly, the metabolism of rabeprazole is largely nonenzymatic and much less dependent on CYP2C19 than other drugs in its class. Because of its lower dependency on the CYP450 enzymatic system, rabeprazole has a greater consistency of pharmacokinetics across a wide range of patients with acid-related disease. CYP2C19 genotype has been demonstrated to affect the pharmacokinetics, pharmacodynamics and therapeutic outcomes of proton pump inhibitors that are primarily dependent on the enzyme for metabolism. The pharmacodynamic profile for rabeprazole is also characterized by more rapid suppression of gastric acid secretion than are other proton pump inhibitors. The favourable pharmacokinetic/pharmacodynamic profile of rabeprazole has proven beneficial for eradication of H. pylori when combined with two antibiotics in both extensive and poor metabolizers. Pharmacodynamic results have also suggested that rabeprazole may be better suited for on-demand therapy in symptomatic GERD patients. Finally, the use of rabeprazole is not complicated by clinically significant drug–drug interactions of the type that have been reported for other proton pump inhibitors.