Twice-daily dosing of esomeprazole effectively inhibits acid secretion in CYP2C19 rapid metabolisers compared with twice-daily omeprazole, rabeprazole or lansoprazole




Twice-daily dosing of proton pump inhibitors (PPIs) is used to treat Helicobacter pylori or acid-related diseases, such as gastro-oesophageal reflux disease (GERD) refractory to standard dose of a PPI. Genetic polymorphisms of CYP2C19 are involved to different extents in the metabolism of four kinds of PPIs (omeprazole, lansoprazole, rabeprazole and esomeprazole) available in Japan.


To compare acid-inhibitory effects of the four PPIs dosed twice daily in relation to CYP2C19 genotype.


We performed 24-h pH monitoring studies on Day 7 of PPI treatment for 40 Japanese H. pylori-negative volunteers [15 CYP2C19 rapid metabolisers (RMs), 15 intermediate metabolisers (IMs) and 10 poor metabolisers (PMs)] using a randomised four-way crossover design: omeprazole 20 mg, esomeprazole 20 mg, lansoprazole 30 mg and rabeprazole 10 mg twice daily.


Although median pH values with esomeprazole, omeprazole, lansoprazole and rabeprazole were 5.7 (3.5–7.2), 5.5 (2.4–7.2), 5.5 (3.7–7.3) and 5.2 (2.5–7.3), respectively (no statistically significant differences), CYP2C19 genotype-dependent differences were smaller for esomeprazole and rabeprazole compared with values for omeprazole and lansoprazole. In CYP2C19 RMs, the median pH with esomeprazole [5.4 (3.5–6.8)] was significantly higher than those with omeprazole [5.0 (2.4–5.9), P = 0.018], lansoprazole [4.7 (3.7-5.5), P = 0.017] or rabeprazole [4.8 (2.5–6.4), P = 0.002]. In IMs and PMs, the median pH was >5.0 independent of the PPI.


In intermediate and rapid metabolisers of CYP2C19, PPIs dosed twice daily could attain sufficient acid suppression, while in CYP2C19 RMs, esomeprazole 20 mg twice daily caused the strongest inhibition of the four PPIs. Therefore, esomeprazole may be effective in Japanese population when dosed twice daily.


Given the importance of rapidly and potently neutralising intragastric pH in the treatment of acid-related diseases such as peptic ulcer or gastro-oesophageal reflux disease (GERD), proton pump inhibitors (PPIs) have enjoyed widespread use in treatment regimens.[1-4]

PPIs undergo extensive hepatic metabolism by the cytochrome P450 (CYP) enzyme system, particularly by CYP2C19.[5] More than 20 variants of the CYP2C19 gene have been discovered, and the pharmacokinetics and pharmacodynamics (i.e. intragastric pH) of PPIs differ by CYP2C19 genotypes.[6-8] The majority of Japanese people can be classified into three genotypes [rapid metaboliser (RM), intermediate metaboliser (IM) and poor metaboliser (PM)] based on the CYP2C19 wild-type gene and two mutated alleles (*2 and *3).[5, 9-11] While a fourth genotype, ultra rapid metaboliser (UM), which is marked by a mutated *17 allele, has also been classified, its incidence in Japan is extremely low.[12] In PMs, plasma PPI levels are markedly increased and acid inhibition by a PPI is enhanced compared with RMs and IMs.[7, 8, 13-15] On the other hand, acid inhibition in RMs is weak compared with IMs and PMs and cure rates of acid-related diseases by PPI-based therapy in RMs are low,[7, 8, 13-15] indicating that CYP2C19 genotype profoundly influences success of PPI treatment. An important point is that prevalence of CYP2C19 genotype status differs among different races: prevalence of CYP2C19 RMs is 56–69% in Caucasians, 81% in African-Americans, 27–35% in Japanese, 38% in Chinese and 13% in Koreans (13%).[5]

Four types of PPIs are currently available in Japan: omeprazole, lansoprazole, rabeprazole and esomeprazole. Of these, omeprazole and lansoprazole are first-generation PPIs and primarily metabolised by CYP2C19 and CYP3A4.[5] Therefore, large differences in the plasma concentrations of omeprazole and lansoprazole are observed among CYP2C19 genotypes, and these differences influence the degree of inhibition of acid secretion.[13] In contrast, the second-generation PPI, rabeprazole, is metabolised mainly via a non-enzymatic pathway with minor CYP2C19 and CYP3A4 involvement.[5] Therefore, the acid-inhibitory activity of rabeprazole is less influenced by the CYP2C19 genotype than omeprazole or lansoprazole.[7] Esomeprazole, the S-isomer of omeprazole, also more effectively inhibits acid secretion than omeprazole,[16, 17] and because of its less extensive first-pass hepatic metabolism than omeprazole, a high plasma level of esomeprazole can be attained and sustained for a longer time than with omeprazole.[17] The increased systemic bioavailability of esomeprazole offers the prospect of improved clinical efficacy and more effective management for acid-related diseases than omeprazole. However, while this pharmacokinetic advantage translates into more effective and more sustained acid inhibition compared with other PPIs in Western populations, no supporting data have been generated in Japanese populations.[18]

Twice-daily dosing of a PPI is used to eliminate Helicobacter pylori or treat patients with GERD refractory to standard dose of a PPI. Because clarithromycin and amoxicillin are acid-sensitive, acid secretion must be potently inhibited by a PPI to prevent the degradation of these antibiotics at low pH.[19] However, previous studies have indicated difficulty in sustaining high pH over 24 h, especially in CYP2C19 RMs, resulting in the eradication rate in RMs being poorest among the three genotypes when receiving first-generation PPIs.[14, 20, 21] Despite this unmet need, however, whether or not twice-daily dosing of esomeprazole or rabeprazole can attain the necessary acid inhibition for H. pylori eradication in CYP2C19 RMs has not been fully elucidated. Similarly, to control of severity and incidence rate of reflux erosive oesophagitis by PPI treatment have been reported to depend on CYP2C19 genotype.[3, 22-25] The control of severity and incidence rate of reflux erosive oesophagitis, intragastric pH must be sustained above 4.0 for 20 h per day;[26] however, as mentioned before, whether or not twice-daily dosing of second-generation PPIs can attain this level of acid inhibition in CYP2C19 RMs is unknown.

Here, to determine whether the acid inhibition attained by twice-daily dosing with esomeprazole and rabeprazole is influenced by CYP2C19 genotype and compare the acid-inhibitory effects of four PPIs dosed twice daily (marketed doses in Japan) in Japanese populations, we conducted a study in healthy Japanese volunteers.

Materials and methods

Study protocol

The inhibition of gastric acid secretion over 24 h by four PPIs (omeprazole, lansoprazole, rabeprazole and esomeprazole) was observed in young Japanese subjects categorised as one of three CYP2C19 genotypes: RM, IM or PM. All subjects gave written informed consent prior to the trial. The study was approved by the Ethics Committee of Hamamatsu University School of Medicine, Hamamatsu, Japan, and was performed in accordance with the ethical principles of the Declaration of Helsinki.

The study was performed as a randomised four-way crossover design. Subjects were dosed a PPI twice daily (at 8:00 AM and at 7:30 PM) for 7 days using the standard dose of H. pylori eradication in Japan as follows: omeprazole 20 mg (Omepral; AstraZeneca, Osaka, Japan), esomeprazole 20 mg (Nexium; AstraZeneca), lansoprazole 30 mg (Takepron; Takeda, Tokyo, Japan) and rabeprazole 10 mg (Pariet; Eisai, Tokyo, Japan). In addition, 40 H. pylori-negative healthy young Japanese individuals without any gastrointestinal symptoms were recruited and gave written informed consent (CYP2C19 genotype; RM, n = 15, IM, n = 15 and PM, n = 10). We excluded subjects with H. pylori infection, a history of H. pylori eradication therapy, a significant clinical illness (e.g. malignancy) or those who were taking any medications [e.g. acid-suppressant drugs, nonsteroidal anti-inflammatory drugs (NSAIDs), antibiotics or calcium channel blockers] from the study.

A 24-h pH monitoring study was performed before the trial as a control and on day 7 of each regimen. At first, an antimony pH catheter was inserted transnasally into all subjects after they were provided a low-fat breakfast; they were then provided two low-fat meals [lunch (591 kcal) at 12:00 PM, dinner (611 kcal) at 7:00 PM] after the electrode monitoring was commenced. Mineral water was allowed ad libitum, but no other beverages were permitted. No subjects consumed alcohol or smoked, and none took any medications from at least 2 weeks before and during the study.

H. pylori status and CYP2C19 genotyping

Helicobacter pylori infection was determined using a serological test (E plate Eiken H. pylori antibody; Eiken Chemical Co. Ltd, Tochigi, Japan). DNA was extracted from leucocytes of a blood sample using a commercially available kit (Wizard Genomic DNA Purification Kit; Promega, Madison, WI, USA). Genotyping procedures for identifying the CYP2C19 wild-type gene and two mutated alleles, CYP2C19 *2 and *3, were performed via the allele-specific primer-polymerase chain reaction (ASP-PCR) assay. All subjects were then classified into three groups, namely RMs, IMs and PMs.

Measurement of intragastric pH

24-h pH monitoring was started at 8:00 AM on Day 7 of each trial. An antimony pH catheter (Synectics Medical, Barcarena, Portugal) linked to a Digitrapper record system (Sierra Scientific Instruments, LLC, Los Angeles, CA, USA) was inserted transnasally under fluoroscopic guidance and placed approximately 5 cm below the lower oesophageal sphincter 30 min before drug administration.

Statistical analysis

Age, height and body weight are expressed as mean ± s.d. Intragastric pH and the percentage time of pH >4 over the 24-h period are expressed as the median (with range). Statistically significant differences among CYP2C19 genotypes in each regimen were determined using the Mann–Whitney U-test if a significant difference was observed using the Kruskal–Wallis test. Statistically significant differences among different regimens were determined using the Wilcoxon signed rank test. All P values are two-sided, and P values <0.05 are considered statistically significant. Calculations were carried out using the statistical software StatView 5.0 (SAS Institute, Cary, NC, USA).


All 40 H. pylori-negative subjects completed the study from October 2011 to December 2012. No subjects experienced any severe adverse events. There were no significant differences in backgrounds among the three CYP2C19 genotype groups (Table 1).

Table 1. Characteristics of enrolled Helicobacter pylori-negative healthy subjects
 Total (n = 40)CYP2C19 RM (n = 15)CYP2C19 IM (n = 15)CYP2C19 PM (n = 10)P value
  1. RM, rapid metaboliser; IM, intermediate metaboliser; PM, poor metaboliser of CYP2C19.

  2. All values are shown as mean ± standard deviation or median (range).

Gender (male: female)30:1012:311:47:30.837
Age (years)21.7 ± 1.722.1 ± 1.521.7 ± 2.021.2 ± 1.80.473
Height (cm)169.7 ± 8.4168.7 ± 10.9167.6 ± 7.0167.2 ± 6.80.853
Weight (kg)59.7 ± 9.960.4 ± 12.259.5 ± 8.158.7 ± 9.30.969
CYP2C19 genotypes *1/*1: n = 15

*1/*2: n = 11

*1/*3: n = 4

*2/*2: n = 5

*2/*3: n = 4

*3/*3: n = 1

Control 24 h intragastric pH1.51 (0.72–3.14)1.55 (0.75–2.51)1.41 (0.80–2.17)1.66 (0.72–3.14)0.614

The median 24-h pH–time curves of each PPI increased compared with controls (nontreatment) and were similar among four kinds of PPIs over 24 h (Figure 1a). The median (range) pH for the control was 1.5 (0.7–3.1), which was significantly lower than those values for esomeprazole [5.7 (3.5–7.2)], omeprazole [5.5 (2.4–7.2)], lansoprazole [5.5 (3.7–7.3)] and rabeprazole [5.2 (2.5–7.3)] (all P < 0.001) (Figure 1b). Similarly, when we compared the percentage time of pH >4 of control for each PPI regimen, statistically significant differences were observed (all P < 0.001) (Figure 1c). However, no significant differences in median pH and percentage time of pH >4 were noted among the four PPIs, except when comparing esomeprazole and rabeprazole (P < 0.001) (Figure 1b,c).

Figure 1.

Median 24-h intragastric pH–time curves for omeprazole (OPZ), lansoprazole (LPZ), rabeprazole (RPZ) and esomeprazole (EPZ) (a) their median intragastric pH values (b) and their percentage time of intragastric pH > 4 (c) *< 0.05.

Although the 24-h pH–time curves for omeprazole and lansoprazole differed among the CYP2C19 genotypes, differences for esomeprazole and rabeprazole between the genotypes appeared smaller (Figure 2a–d). The 24-h median (range) pH for esomeprazole in PMs was 6.2 (5.2–7.2), which was significantly higher than those values in RMs [5.4 (3.5–6.8), P = 0.018] or IMs [5.6 (4.3–6.8), P = 0.017] (Figure 2e). When omeprazole or lansoprazole were dosed, significant differences in 24-h median pH values were observed among CYP2C19 genotypes [omeprazole: RMs, 5.0 (2.4–5.9), IMs, 5.7 (4.5–6.5) and PMs, 6.6 (5.5–7.2); and lansoprazole: RMs, 4.7 (3.7–5.5), IMs, 5.4 (4.7–6.9) and PMs, 6.4 (5.5–7.3); P < 0.05] (Figure 2e). Significant differences were noted between RMs and PMs and between IMs and PMs in percentage time of pH >4 for esomeprazole and omeprazole (P < 0.05 respectively) (Figure 2f). For rabeprazole, the median pH and percentage time of pH >4 significantly differed only between RMs and PMs (Figure 2e,f). These observations suggest that acid inhibition by the twice-daily dosing of any of the PPIs, including esomeprazole, differed among the CYP2C19 genotypes in H. pylori-negative Japanese subjects.

Figure 2.

The 24-h pH–time curves of omeprazole (a) lansoprazole (b) rabeprazole (c) and esomeprazole (d) their median intragastric pH values (e) and the percentage time of intragastric pH > 4 (f) in three different CYP2C19 genotypes. *P < 0.05, #P < 0.001. EPZ, esomeprazole; I, CYP2C19 intermediate metaboliser; LPZ, lansoprazole; OPZ, omeprazole; PM, CYP2C19 poor metaboliser; RPZ, rabeprazole; R, CYP2C19 rapid metaboliser.

In CYP2C19 RMs, the median (range) pH of esomeprazole [5.4 (3.5–6.8)] was significantly higher than that of omeprazole [5.0 (2.4–5.9), P = 0.018], lansoprazole [4.7 (3.7–5.5), P = 0.017] or rabeprazole [4.8 (2.5–6.4), P = 0.002] (Figure 3a). Esomeprazole alone attained an intragastric pH higher than 5.0 in RMs. In PMs, the median (range) percentage time of pH >4 with omeprazole [94.5% (84.8–100%)] was significantly higher than that with esomeprazole [87.2% (71.6–99.2%), P = 0.017] or rabeprazole [86.6% (21.0–99.2%), P = 0.047] (Figure 3b). However, the median of intragastric pH was higher than 5.0 in all IMs and PMs independent of the PPI.

Figure 3.

Comparisons of median intragastric pH values (a) and percentage time of intragastric pH > 4 (b) among the four PPIs. *P < 0.05. EPZ, esomeprazole; IM, CYP2C19 intermediate metaboliser; LPZ, lansoprazole; OPZ, omeprazole; PM, CYP2C19 poor metaboliser; RPZ, rabeprazole; RM, CYP2C19 rapid metaboliser.


We demonstrated that the 24-h pH values attained overall with twice-daily dosing of PPIs were equivalent among the four kinds of PPIs available in Japan. However, the variations in acid inhibition associated with the CYP2C19 genotype were small for esomeprazole and rabeprazole compared with values for omeprazole and lansoprazole. In H. pylori-negative Japanese subjects with CYP2C19 RMs, median pH was significantly higher with esomeprazole than with the other three PPIs. Although median pH in PMs was significantly higher with omeprazole than with other PPIs, including esomeprazole, acid inhibition attained by any of the four PPIs in IMs and PMs was sufficient for treatment of GERD and H. pylori eradication.[4, 26-28] Therefore, although influence of CYP2C19 genotype status on acid inhibition cannot be ignored, we deem esomeprazole to be preferable PPI for use in treating acid-related diseases, particularly in Japanese with CYP2C19 RMs.

In RMs, PPIs were rapidly eliminated from the systemic circulation, and plasma PPI levels before and 3 h after dosing were often below detectable levels, resulting in insufficient acid inhibition. Following the rapid elimination of PPIs, newly generated or activated H+, K+-ATPases in gastric parietal cells are able to secrete gastric acid.[6-8, 13] However, acid inhibition by PPIs has also been observed to occasionally be insufficient, even if dosed twice daily in CYP2C19 RMs.[8] Because the pharmacokinetics and pharmacodynamics of PPIs differ by the dependence on CYP2C19 in each metabolic pathway,[7, 8, 13-15, 29] PPI with less influence of CYP2C19 is more preferable. One attractive characteristic of esomeprazole is that it is metabolised by CYP2C19 to a lesser extent than omeprazole and therefore achieves potent acid inhibition over 24 h, irrespective of CYP2C19 genotype.[16, 17, 30] However, because CYP3A4 is also involved in the metabolism of esomeprazole, esomeprazole has the potential risk of drug–drug interaction with a variety of drugs metabolised by CYP3A4.[31] Physicians must pay attention to this problem.

In a randomised crossover study using healthy H. pylori-negative subjects, the percentage time of pH >4 with esomeprazole 20 mg was significantly longer than that with lansoprazole 15 mg (50.4% vs. 43.0%, P = 0.03) or rabeprazole 10 mg (59.8% vs. 51.7%, P = 0.01).[32] In the present study, the median 24-h pH with esomeprazole 20 mg twice daily in H. pylori-negative CYP2C19 RMs was significantly higher than that with omeprazole, rabeprazole or lansoprazole throughout the 24-h observation period. We therefore recommend the use of esomeprazole in patients with acid-related diseases, particularly those refractory to standard doses of a PPI (e.g. CYP2C19 RMs).

Several studies have examined the efficacy of twice-daily dosing of a PPI with relation to CYP2C19, with conflicting findings being reported. For example, while two studies noted CYP2C19-dependent differences in acid inhibition between lansoprazole 30 mg and rabeprazole 20 mg, no such differences were seen between rabeprazole 10 mg and esomeprazole 20 mg.[33, 34] In the present study, the intragastric pH attained by twice-daily dosing of PPIs differed among CYP2C19 genotypes. However, the CYP2C19 genotype-dependent variance in intragastric pH attained by esomeprazole and rabeprazole was smaller than that attained by lansoprazole or omeprazole. Therefore, twice-daily dosing of second-generation PPIs might overcome the influence of CYP2C19 genotypes to some degree, although not completely. As an alternative, trials of four-times daily dosing of PPIs have also been reported, involving rabeprazole 10 mg, lansoprazole 30 mg and esomeprazole 10 mg, all of which attained sufficient acid inhibition in CYP2C19 RMs.[8, 34, 35] Taken together, these present and previous findings therefore suggest that multiple dosing of a PPI decreases CYP2C19 genotype-dependent differences in acid inhibition.

Potent acid inhibition is important for the eradication of H. pylori.[4] Potent acid inhibition also increases the stability and bioavailability of antibiotics in the stomach and the concentration of antibiotics in gastric mucosa.[36-38] Furthermore, acid inhibition allows H. pylori to reach its growth phase, rendering the bacteria more sensitive to antibiotics.[38] We [28] previously reported that, in analysis of 24-h pH monitoring using H. pylori-negative subjects, 24-h pH level was required to be higher than 5.0, which was achieved in CYP2C19 IMs with eradication rate of higher than 95%. In addition, we demonstrated that the pH level over 24 h was significantly higher in patients who achieved successful eradication using lansoprazole-based triple therapy [6.4 (5.0–7.6)] than those who did not [5.2 (2.2–6.2)].[4]

In this study, median pH with esomeprazole in H. pylori-negative CYP2C19 RMs was [5.4 (3.5–6.8)], but those attained with other PPIs were not higher than 5.0, suggesting that esomeprazole-based eradication might be acceptable compared with other PPIs for CYP2C19 RMs. We therefore will plan to show in our hypothesis that H. pylori eradication rate with esomeprazole-based regimen is independent of the CYP2C19 genotype and is higher than that with an omeprazole-, rabeprazole- or lansoprazole-based regimen in CYP2C19 RM patients.

PPIs improve acid reflux heartburn symptoms and oesophageal mucosal breaks.[39-43] Meta-analyses of treatments for erosive GERD patients have shown that PPIs are much more effective in curing oesophageal erosions and acid reflux-related symptoms than H2 receptor antagonists.[44, 45] Furthermore, twice-daily dosing of rabeprazole was effective in treating patients with GERD refractory to standard doses of PPI therapy.[46] However, improvements in heartburn associated with non-erosive reflux disease (NERD) using standard PPI dosages are lower (ranging from 30 to 60%) than those for erosive GERD (ranging from 10 to 30%).[47-49] The development of such oesophageal mucosal damage is closely related to intraoesophageal and intragastric pH values. For effective treatment of GERD using acid-inhibitory drugs, intragastric pH over a 24-h period should fall below <4.0 for no longer than 2–4 h.[26] In this study, in RMs, the median percentage time of pH <4 ranged from 26.6 to 38.7% independent of the PPI used, suggesting that PPI twice-daily dosing may be insufficient for treating GERD/NERD patients refractory to standard doses of a PPI.

Several limitations to the present study warrant mention. First, the subjects were all young, healthy, H. pylori-negative volunteers, not patients. Because acid secretion and response to PPIs depend on age, the grade of gastric mucosal atrophy and H. pylori infection, our results should be extrapolated with caution to patients treated with a PPI for H. pylori eradication or GERD. Second, we did not measure the plasma PPI levels; therefore, we could not directly correlate the metabolic disposition of each PPI with the pH attainted with each PPI. Third, the sample size was small, particularly when stratifying data based on CYP2C19 genotype group. Enrolment of a large population of H. pylori-positive patients will be required to show significant evidence that the cure rate of patients treated with esomeprazole-based H. pylori eradication or patients refractory to standard PPI treatment is better than that of therapy based on another PPI. Lastly, we investigated acid-inhibitory effects at twice-daily dosing of rabeprazole 10 mg (a marketed dose in Japan), not 20 mg (a standard dose worldwide). Therefore, the present study results cannot be extrapolated to Western patients. In general, because a dose-dependent increase in median pH and in percentage time of pH > 4 was observed for rabeprazole administered once daily in doses of 5, 10, 20 and 40 mg,[50] further study to compare the efficacy for acid inhibition attained with twice-daily dosing of rabeprazole 20 mg and esomeprazole 20 mg will be required.

In conclusion, we demonstrated that, for Japanese people, esomeprazole is more effective than other PPIs tested for achieving potent acid inhibition when dosed twice daily in CYP2C19 RMs who are at risk of being unresponsive to PPI therapy at the standard dose. The clinical potential of twice-daily dosing of esomeprazole must be verified in future studies under appropriate protocols.


Guarantor of the article: None.

Author contributions: All authors approved the final version of the manuscript.


Declaration of personal interests: None.

Declaration of funding interests: Grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan (23590912 and 23590913) and from Japan Research Foundation for Clinical Pharmacology. First Department of Medicine and The Center for Clinical Research at Hamamatsu University School of Medicine have received grants from Takeda Pharmaceutical Co., Ltd., AstraZeneca KK, Eisai Co., Ltd., Daiichi-Sankyo Co. Ltd.