Anti-platelet drug-induced gastric mucosal injury correlates with intragastric pH. Our aim was to investigate prophylaxis effects of famotidine, one of the representative histamine-2 receptor antagonists (H2RA), on gastric mucosal injury induced by dual therapy with low-dose aspirin and clopidogrel in relation to Helicobacter pylori (H. pylori) infection and CYP2C19 genotypes. This study was conducted for 20 healthy Japanese volunteers (10 H. pylori-positive and 10-negative subjects) with 100 mg aspirin plus 75 mg clopidogrel (AC) once-daily dosing and AC plus 20 mg famotidine twice-daily dosing (ACH). Mucosal injury was endoscopically assessed on day 3 and 7 and 24-hour intragastric pH and antiplatelet-function test was performed on day 7. Median pH in ACH was similar between CYP2C19 extensive metabolizer (EM) and intermediate metabolizer (IM)/poor metabolizer (PM) and was significantly higher in H. pylori-positive than negative subjects (P < .05). Mucosal injury with ACH significantly decreased in both day 3 and 7 compared with AC, irrespective with H. pylori and CYP2C19 genotypes (P < .05). Although antiplatelet effect of clopidogrel in EM was significantly higher than that in IM/PM, the additional famotidine did not affect the effect. Anti-platelet drug-induced gastric injury was alleviated by famotidine without attenuation of anti-platelet functions irrespective of H. pylori and CYP2C19 genotypes.
aspirin reaction unit
- H. pylori
histamine-2 receptor antagonist
inhibition of platelet aggregation
modified Lanza score
non-steroidal anti-inflammatory drugs
proton pump inhibitor
Verify Now system
The benefits of dual antiplatelet therapy with low-dose aspirin and clopidogrel have been established in the treatment and prevention of cerebrovascular diseases and ischemic heart diseases.[1, 2] However, low-dose aspirin increases the risk of not only gastrointestinal mucosal injury but also gastrointestinal bleeding.[3-8] Clopidogrel also causes gastric mucosal bleeding. Previously, we showed that low-dose aspirin, clopidogrel or both induced gastric mucosal injury in more than 90% of young, healthy, Japanese Helicobacter pylori (H. pylori)-negative volunteers.[4, 6, 9, 10] In these studies, clopidogrel was associated with an increase in hemorrhage score rather than erosion score in terms of modified Lanza score.[6, 9, 10]
The incidence and severity of low-dose aspirin-induced gastric mucosal injury and bleeding have significant correlations with intragastric pH during low-dose aspirin treatment.[4-6] Therefore, the use of a potent acid inhibitor, such as a proton pump inhibitor (PPI) or a histamine-2 receptor antagonist (H2RA), is recommended for the prevention and treatment of gastric mucosal injury induced by low-dose aspirin. However, because acid inhibition with an H2RA is weak compared with that with a PPI,[5, 12, 13] an H2RA was reported to be inferior to a PPI for the prevention of low-dose aspirin-induced gastric mucosal injury.
PPIs are mainly metabolized by CYP2C19 in the liver. There are genetic differences in the activity of this enzyme. In individuals with the extensive metabolizer (EM) genotype of CYP2C19, a PPI is sooner metabolized and eliminated from the systemic circulation than in intermediate metabolizers (IM) and poor metabolizers (PMs), resulting that CYP2C19 EMs have insufficient acid suppression during PPI treatment.[6, 12, 15, 16] This CYP2C19 EM genotype-related insufficient acid inhibition by a PPI results in insufficient clinical outcomes of PPI-based therapies, such as treatment for GERD and eradication of H. pylori.[6, 17, 18] On the other hand, because famotidine, one of the representative H2RAs, is eliminated via the kidney, the activity of famotidine is not influenced by CYP2C19 genotypes.
Clopidogrel is an inactive prodrug requiring several biotransformation steps, mediated mainly by cytochrome P-450 isoforms (CYP), in order to generate an active metabolite with a thiol moiety that binds irreversibly to the platelet ADP receptor P2Y12. The conversion of clopidogrel to the active metabolite is required to be a two-step, CYP-dependent process. In these steps, CYP2C19, 1A2, 2B6, 2C9, and 3A4/5 may be involved in the activation of clopidogrel.[19, 20] Of CYP enzymes, recently, the antiplatelet activity of clopidogrel was reported to depend on CYP2C19 genotypes, as well as other CYP enzymes.[19, 20] On the other hand, famotidine does not attenuate the active metabolism of clopidogrel because famotidine does not influence the activity of CYP2C1921. However, the prophylactic effect of famotidine on the gastric mucosal injuries induced by dual antiplatelet therapy has not been fully elucidated.
Infection by H. pylori causes gastric mucosal inflammation and influences acid secretion by gastric mucosal atrophy and up-regulation of pro-inflammatory cytokines.[22, 23] H. pylori infection is also associated with the incidence and severity of drug-related gastric mucosal injury. However, it is unclear whether the preventive effects of H2RA on dual antiplatelet therapy-induced gastric mucosal injury depend on H. pylori infection.
Given the above information, we aimed to investigate usefulness of H2RA as prophylaxis for gastric mucosal injury induced by dual anti-platelet therapy with low-dose aspirin and clopidogrel in relation to H. pylori infection and CYP2C19 genotypes.
Materials and Methods
Approval for the study protocol was given in advance by the Human Institutional Review Board of the Hamamatsu University School of Medicine and we obtained a written informed consent from patients enrolled in this study.
We recruited 118 healthy Japanese volunteers, after obtaining their written informed consent (Figure 1). Exclusion criteria were: past treatment for H. pylori infection, past history of peptic ulcer or abdominal operation, regular use of nonsteroidal anti-inflammatory drugs (NSAIDs), low-dose aspirin, clopidogrel, or any other medications, smoking, and volunteers who were less than 20 years old. After checking for H. pylori infection and CYP2C19 genotypes, we randomly selected 10 H. pylori-negative and 10 H. pylori-positive subjects in both groups (5 EMs and 5 IM/PMs), as shown in our previous report (Figure 1).
All 20 subjects underwent gastroscopy, 24-hour intragastric pH monitoring and platelet function tests before this study to exclude gastroduodenal disorders, abnormal gastric acid secretion and abnormal platelet function. All subjects were found to have normal gastric acid secretion and normal platelet function without any gastroduodenal disorders.
Previously, we performed a randomized, open-labeled, four-way cross-over study including aspirin alone, clopidogrel alone, combination of aspirin and clopidogrel, and combination of aspirin/clopidogrel and rabeprazole. After this study, as 5th regimen, we additionally performed ACH regimen, aspirin at 100 mg once-daily (Bayaspirin®, enteric-coated aspirin tablet, Bayer Schering Pharma, Osaka, Japan), clopidogrel at 75 mg once-daily (Plavix®, Sanofi-Aventis K.K., Tokyo, Japan) plus famotidine at 20 mg twice daily (Gaster®, Astellas Pharma, Inc., Tokyo, Japan) for same 20 subjects.
Aspirin and clopidogrel were dosed once daily at 8:00 am and famotidine 20 mg was dosed twice daily at 8:00 am and 6:00 pm for 7 days. Subjects were not permitted to consume alcohol or take any other medications for at least 2 weeks prior to each regimen period and during the study period. All subjects were provided with low-fat meals (lunch at 12:30 pm and dinner at 6:00 pm) on day 7. Mineral water was allowed ad libitum, but no other beverages were permitted. The washout period between the regimens was at least 2 weeks. Gastroscopy and platelet function testing were performed on days 3 and 7 of each regimen. Twenty-four-hour intragastric pH monitoring was performed on day 7.
CYP2C19 Genotyping and H. pylori Infection
DNA was extracted from each subject's leukocytes using a commercially available kit (Wizard® Genomic DNA Purification Kit, Promega Co., Madison, WI, USA). Genotyping for identifying the CYP2C19 wild-type gene (1*) and the mutated alleles, CYP2C19*2 in exon 5 and CYP2C19*3 in exon 4, was performed using polymerase chain reaction (PCR) restriction based on the Q probe method (Genecube®; Toyobo Co., Ltd., Osaka, Japan). Subjects were classified into three genotype groups: EM (*1/*1), IM (*1/*2 or *1/*3) and PM (*2/*2, *2/*3, or *3/*3).
Screening for H. pylori infection was conducted using a serological test (E plate Eiken H. pylori antibody®, Eiken Chemical Co. Ltd., Tochigi, Japan).
After an overnight fast, gastroscopy was performed using an Olympus GIF-Q 260® flexible gastroscope (Olympus Co., Tokyo, Japan). Gastric mucosal damage was graded using the MLS system by three experienced endoscopists blinded to information about the subjects.[9, 10] The average scores by the three endoscopists were used as the MLS for each subject.
Twenty-Four-Hour Intragastric pH Monitoring
Twenty-four-hour intragastric pH monitoring was started 30 minutes after gastroscopy. A single crystal antimony multiuse pH catheter® (Synectics Medical, Barcarena, Portugal) was inserted transnasally and placed at 5 cm distal to the gastric cardia. We confirmed radiographically the placement of pH probe. Recorded data were analyzed using a Digitrapper pH400® (Sierra Scientific Instruments LLC, Los Angeles, CA, USA).
Platelet Function Testing for Aspirin and Clopidogrel
We measured the inhibition of platelet aggregation (IPA) by aspirin and clopidogrel [Verify Now system (VN) Aspirin assay® and VN P2Y12 assay®, respectively, Accumetrics, Inc., CA, USA] at 8:00 am. We calculated IPA for low-dose aspirin using the value of Aspirin Reaction Units (ARU) of the VN Aspirin assay® as follows: IPA = [(ARU control − postdose)/ARU control] × 100%. For IPA by clopidogrel, the data of the VerifyNow P2Y12 assay® were used.
MLS, pH, and platelet function values are expressed as median and range. Other values such as age, height and weight are given as mean and SD. Wilcoxon's signed-rank test, Friedman's test, the Mann–Whitney U-test and the Kruskal–Wallis test were used as appropriate for statistical analysis using SPSS (IBM® SPSS Statistics, Ver. 20, IBM, Armonk, NY, USA). All P-values were two-sided, and P < .05 was considered as indicating statistical significance.
No significant differences in age [H. pylori-positives (21.8 ± 2.0 years old) and negatives (21.8 ± 2.1 years old)], sex [number, male:female; 7:3 (H. pylori-positives) =7:3 (H. pylori-negatives)], height [170.0 ± 9.0 and 170.1 ± 11.9 cm] and weight (62.6 ± 12.0 and 65.5 ± 11.1 kg) were observed in H. pylori-positive and -negative subjects and the results were similar with different CYP2C19 genotypes. The mean serum pepsinogen I level in H. pylori-positive subjects (71.5 ± 23.2 ng/mL) was significantly higher than that in -negative subjects (40.5 ± 6.3, P < .05). The pepsinogen I/II ratio in H. pylori-positive subjects (2.9 ± 0.8) was significantly lower than that in -negative subjects (4.7 ± 0.7, P < .05).
Gastric Mucosal Damage and 24-Hour Intragastric pH
The median 24-hour intragastric pH significantly differed between the AC [1.7 (1.0–4.3)] and ACH [3.6 (1.8–7.4)] regimens on day 7 (P < .01). The median pH in the ACH regimen was significantly higher than that in the AC regimen (Figure 2A). Similar differences were observed in the median percent time with pH < 4 between two regimens—ACH: 61.8% (0–91.6%) and AC 90.6% (50.0–99.5%)] (Figure 2B).
The AC regimen for only 3 days induced moderate mucosal injury [MLS: 2.8 (0–4.0)], which was sustained until day 7 (Figure 2C). The MLSs in the ACH [1.3 (0–3.3) and 1.0 (0–3.3)] regimens were significantly lower than those of the AC regimen [2.8 (0–4.0) and 2.7 (0.3–4.0)] on days 3 and 7 (all P < .01) (Figure 2C).
Gastric Mucosal Damage and 24-Hour Intragastric pH From the Viewpoint of H. pylori Infection
When we compared intragastric pH in H. pylori-positive subjects with that of -negative subjects, the median pHs in H. pylori-positive subjects in the ACH regimen were significantly higher than those of -negative subjects (P < 0.05 and 0.01, respectively) (Figure 3A). This observation suggested that acid inhibition attained with acid inhibitory agents depended on the presence of H. pylori infection.
The ACH regimens significantly decreased the medians of MLSs in comparison with the AC regimen, irrespective of H. pylori infection (Figure 3B). However, no significant differences in MLSs were observed between H. pylori-negative and -positive subjects in ACH regimen (Figure 3B).
Gastric Mucosal Damage and 24-Hour Intragastric pH From the Viewpoint of CYP2C19 Genotypes
CYP2C19 genotype-dependent difference in the median pH was not observed in the ACH regimen (Figure 4A). Gastric mucosal injury was effectively prevented by the ACH regimens in not only IMs/PMs, but also EMs (Figure 4B).
Platelet Function Testing
The IPA of low-dose aspirin had no significant differences between the two CYP2C19 genotype groups (EMs vs. IMs/PMs) (data not shown). On the other hand, the IPA of clopidogrel in EMs on day 7 was significantly higher than that in the IM/PM group. The addition of H2RA did not influence the IPA of clopidogrel (Table 1).
|Regimen||Control||Day 3||Day 7|
|Total||2 (0–11)||35 (9–78)||43.5 (11–84)||48 (13–89)||52 (16–80)|
|CYP2C19 EM||0 (0–9)||53 (24–78)||63.5 (37–84)||59 (35–89)||63.5 (43–80)|
|CYP2C19 IM/PM||0 (0–11)||30 (9–64)||32.5§ (11–50)||34b (13–57)||47.5a(16–60)|
We studied the effects of H2RA (eg, famotidine) on dual anti-platelet therapy-induced gastric mucosal injury in relation to intragastric pH level, CYP2C19 genotypes and H. pylori infection in a prospective manner, and showed that the preventive effect of an H2RA on gastric injury depended on gastric acidity, but not the presence or absence of H. pylori infection and CYP2C19 genotypes. Therefore, when patients with cerebrovascular or ischemic heart disease are treated by dual antiplatelet therapy with low-dose aspirin and clopidogrel, famotidine could be one of the useful prophylaxis agents.
Rapid and potent neutralization of intragastric pH is effective for the treatment of acid-related diseases such as peptic ulcers, gastroesophageal reflux disease (GERD), and H. pylori eradication therapy. Similarly, the incidence and severity of low-dose aspirin-induced esophageal and gastric mucosal injuries are associated with intragastric pH.[4, 7] Therefore, acid inhibition is necessary to prevent low-dose aspirin-induced gastric injury.[4-6] Previously, we showed that median 24-hour intragastric pH > 5.0 and median percent time with pH < 4.0 less than 40% were required for the prevention of low-dose aspirin-induced esophageal mucosal injury. However, there was no data concerning appropriate acid inhibition to prevent gastric mucosal damage caused by dual anti-platelet therapy, low-dose aspirin and clopidogrel. When we combined data of ACH used in this study and aspirin/clopidogrel/rabeprazole used in our previous study, the 24-hour pHs were significantly correlated with the MLSs in all subjects (P < .01, ρ = −0.46) and maintenance of median 24-hour pH ≥ 5.0 and median percent time with pH < 4.0 less than 40% prevented dual anti-platelet therapy-induced gastric injury. If the subjects attain potent acid inhibition by acid inhibitors, the majority of them could avoid gastric injury. Therefore, we recommend maintaining the intragastric pH higher for a longer time in order to prevent gastric mucosal injury during dual antiplatelet therapy.
Recently, the FAMOUS trial showed that famotidine effectively prevented low-dose aspirin-induced gastric mucosal injury. However, it is unknown whether H2RA treatment is effective for clopidogrel-induced or dual antiplatelet drug-induced gastric mucosal injury. In general, because most H2RAs, such as famotidine, ranitidine and nizatidine, are renally excreted and hepatic enzymes, such as CYP2C19, do not metabolize such H2RAs, no interactions with clopidogrel and H2RAs have been reported. Therefore, an H2RA may be an appropriate alternative for patients treated with clopidogrel to avoid drug–drug interaction. In this study, famotidine treatment effectively prevented low-dose aspirin/clopidogrel-induced gastric mucosal injury, irrespective of CYP2C19 genotypes, suggesting that acid inhibitory effects attained with a standard dose of famotidine (ie, twice daily dosing of 20 mg) might be useful to prevent mucosal injury. Recent evidence showed that PPI effectively prevented low-dose aspirin-induced gastric mucosal injury compared with an H2RA treatment. When we compared with data of ACH used in this study and aspirin/clopidogrel/rabeprazole used in our previous study, no significant differences in acid inhibition and gastric mucosal injury in CYP2C19 EMs, not IM/PMs, were observed between rabeprazole at 10 mg once daily and famotidine at 20 mg twice daily. The standard dose of a PPI could not sustain intragastric pH at more than 5 throughout 24 hours in CYP2C19 EM patients.[27-30] This observation might suggest that famotidine could be replaced with a PPI in CYP2C19 EMs.
Problems concerning the interaction between PPI and clopidogrel remain.[31-33] There have been several in vivo studies showing that the antiplatelet function of clopidogrel is attenuated with PPIs.[33, 34] In addition, retrospective clinical studies have also demonstrated the increased risk of cardiovascular diseases by interaction of PPI and clopidogrel.[32, 35, 36] The US Food and Drug Administration (FDA) also warned that the anti-platelet function of clopidogrel could be impaired by concomitant medications that inhibit CYP2C19 activity, including PPI. However, this study showed that the addition of famotidine did not affect the IPA of clopidogrel and there was no drug–drug interaction of clopidogrel and famotidine. Although prospective studies have shown that combination therapy with PPI and clopidogrel does not increase cardiovascular risk but decreases the incidence of gastroduodenal bleeding as an important clinical outcome, we have to keep in mind that PPI interacts with clopidogrel.
Subjects receiving low-dose aspirin developed gastric mucosal injury within 3 days. It has been reported that gastric mucosal injury by low-dose aspirin improves after 7 days by the adaptive cytoprotection in H. pylori-negative subjects, but not in H. pylori-positive subjects.[38, 39] In general, gastric adaptation against low-dose aspirin was considered to be mediated by increased mucosal cell proliferation. The mucosal gene expression for spasmolytic peptide (a member of trefoil peptides) and transforming growth factor alpha (TGF-alpha) as well as for cyclooxygenase (COX)-1 and COX-2 were decreased on day 3 and significantly restored on day 7 in H. pylori-negative subjects taking low-dose aspirin, while this adaptation was substantially attenuated in H. pylori-positive subjects. However, in the present study, we did not find such gastric adaptation and observed that low-dose aspirin- and clopidogrel-induced mucosal injury was sustained throughout 7 days. Although the detailed mechanism why mucosal injury was sustained throughout 7 days was unclear, we suppose that the attenuation of antiplatelet function by clopidogrel may prevent gastric adaptation to low-dose aspirin. Further study to investigate mucosal gene expression for spasmolytic peptide, TGF-alpha, COX-1, and COX-2 will be required to clarify the mechanism in detail.
The conversion into 2-oxo clopidogrel from clopidogrel is regulated by CYP2C19, CYP2B6 and CYP1A2, and the final transformation into the active metabolite is regulated by CYP2C19, CYP2C9, CYP2B6, CYP3A isozymes. Therefore, CYP3A isozymes also play an important role in the activation of clopidogrel. In fact, clinical consequences of potential drug–drug interactions of clopidogrel with drugs affecting CYP3A isozymes activity was shown in patients receiving concomitant CYP3A isozymes inhibitor, such as atorvastatin, and clopidogrel, the antithrombotic effect of clopidogrel was moderately attenuated, but the combination significantly reduced the overall mortality.
This study has the following limitations. Firstly, the subjects were young healthy volunteers, not elderly patients taking low-dose aspirin and clopidogrel. Secondly, the sample size was small. Thirdly, the distribution of CYP2C19 genotypes in the enrolled subjects was different from the usual distribution of CYP2C19 genotypes in Japan. Fourthly, we could not avoid the influences of treatment courses which may have impacted the assessment of gastric damage. Finally, the study period was too short. Accordingly, our results should be considered preliminary in nature. Further studies with larger patient numbers of older age and of longer duration are required.
In conclusion, we demonstrated that famotidine is effective for the prevention of dual anti-platelet therapy-induced gastric mucosal injury and bleeding, irrespective of H. pylori infection and CYP2C19 genotypes. Although a PPI rather than an H2RA is usually recommended, which effectively prevented gastric mucosal damage induced by dual therapy with low-dose aspirin and clopidogrel, the present study demonstrated that famotidine treatment for combination of clopidogrel and low-dose aspirin effectively prevented gastric mucosal injury without attenuation of anti-platelet functions. In view of this efficacy, and less expensive cost, we have to investigate the clinical usefulness of famotidine in future prospective studies in a larger population.
Declaration of Conflicting Interests
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., and Daiichi-Sankyo Co., Ltd. The authors have no other conflicts of interest that are directly relevant to the content of this article.
This work was supported by 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.