CLINICAL THERAPEUTICSVVOL. 22, NO.

3, 2000

The Proton-Pump Inhibitors: Similarities and Differences

John Horn, PharmD

University of Washington School of Pharmacy, Seattle, Washington

ABSTRACT

Objective: This paper examines the clinical pharmacology of the proton-pump in-
hibitors (PPIs) and briefly reviews some comparative studies of these agents.
Background: PPIs have emerged as the treatment of choice for acid-related diseases, in-
cluding gastroesophageal reflux disease (GERD) and peptic ulcer disease. Although these
drugs+meprazole, lansoprazole, pantoprazole, and rabeprazole-share a common struc-
ture (all are substituted benzimidazoles) and mode of action (inhibition of H’,K+-adeno-
sine triphosphatase [ATPase]), each differs somewhat in its clinical pharmacology.
Results: In comparative clinical trials found in MEDLINE@, PPIs administered once
daily produced endoscopic evidence of healing in >90% of patients with duodenal ulcer
after 4 weeks of treatment, in >90% of those with gastric ulcer after 6 weeks of treatment,
and in ~90% of those with ulcerative or erosive GERD after 8 weeks of treatment. Main-
tenance therapy with daily doses of a PPI has been shown to be an effective means of
preventing GERD relapse. PPIs also inhibit the growth of Helicobacter pylori, now rec-
ognized as an important factor in peptic ulcer disease, and, when administered in combi-
nation with antibiotics, provide the best treatment for eradication of the bacterium.
Rabeprazole has a more rapid onset of H+,K+-ATPase inhibition than the other PPIs and,
compared with omeprazole, a greater effect on intragastric pH after the first dose. Omepra-
zole and lansoprazole have a greater potential for drug-drug interactions than do panto-
prazole and rabeprazole.
Conclusion: Although the individual PPIs have similar efficacy in many cases, differ-
ences between them should be considered when choosing a treatment regimen.
Key words: acid-related disease, gastroesophageal reflux disease, peptic ulcer disease,
proton-pump inhibitors. (Clin Ther: 2000;22:266-280)

Accepted for publication February 2, 2000.

Printed in the USA. Reproduction in whole or part is not permitted.

266 0149.2918/00/$19.00
J. HORN

INTRODUCTION

Proton-pump inhibitors (PPIs) have emerged

as the drug class of choice for treating
patients with acid-related diseases, in-
cluding gastroesophageal reflux disease A
(GERD), duodenal ulcer, and gastric ul- Pantoprazole
cer. PPIs are also effective in treating
patients with Barrett’s esophagus and
Zollinger-Ellison syndrome. CHz CH,
These agents inhibit gastric acid secre-
tion by targeting the gastric acid pump,
H+,K+-adenosine triphosphatase (ATPase),
ti
in the canalicular membrane of the parietal
Omeprazole
cell.‘12The regulation of acid secretion is a
complex process involving many cell
types, hormones, and mediators, but these
processes converge in a final common step
involving H+,K+-ATPase.1As a result, PPIs
effectively inhibit acid secretion in a man-
ner independent of the processes that stim- ii
ulate the parietal cell.’ Lansoprazole
Each member of the PPI class-
omeprazole, lansoprazole, pantoprazole,
Na
and rabeprazole-has been shown to be
effective and well tolerated in random-
ized, controlled clinical trials. This paper
p&fH;~-cT;H*-&cH~
examines the clinical pharmacology of the
N-
PPIs and briefly reviews some compara-
tive PPI studies found in MEDLINE@. Rabeprazole

Figure 1. Chemical structures of the

MECHANISM OF ACTION proton-pump inhibitors.

PPIs are substituted 2-pyridyl methyl-

sulfinyl benzimidazoles that share a simi- environment, protonation of the pyridine
lar core structure (Figure 1). These agents and benzimidazole nitrogens results in for-
are protonatable weak bases with pKa (neg- mation of a tetracyclic sulfenamide, which
ative logarithm of the acid-ionization con- represents the active form of the drug.
stant) values of -4, with the exception of The sulfenamide binds to exposed cys-
rabeprazole, which has a pK, value of 5.t** teine residues in the alpha subunit of
PPIs therefore accumulate selectively in H+,K+-ATPase to form covalent disulfide
acidic spaces with a pH of <4, which are bonds, which inhibit the activity of the
found primarily in the secretory canalicu- pump. Because PPIs bind covalently, the
lus of the gastric parietal cell. In this acidic duration of action of these drugs extends

267
 CLINICAL THERAPEUTICS@

beyond their plasma half-life of -1 to 2 cance of these distinct patterns of binding

hours.’ Gastric acid secretion is restored remains to be determined, but they may
by translocation of new H+,K+-ATPase to help explain the subtle pharmacologic dif-
the secretory canaliculus membrane. ferences between PPIs.
However, some enzyme activity may be
restored, depending on whether glu-
INHIBITION OF GASTRIC ACID
tathione can reverse the disulfide bond.3
SECRETION
Although all PPIs have the same gen-
eral mechanism of action, the nature of the Gastric acid is a critical factor in general
pyridine and benzimidazole substituents gastrointestinal health. Although the exact
may confer physical and chemical differ- nature of its role varies in GERD, gastric
ences. Because of its higher pK,, rabepra- ulcer, duodenal ulcer, Zollinger-Ellison
zole is activated over a wider pH range syndrome, and Barrett’s esophagus, all are
than omeprazole, lansoprazole, or panto- acid-related disorders. Evidence of the im-
prazole, and it converts to its sulfenamide portant role of gastric acid in these condi-
faster than the other 3.4 This profile might tions is provided by the success of antise-
explain why rabeprazole has a faster onset cretory therapy in many patients6
of inhibitory action with respect to H+,K+- Gastric acid activates pepsinogens to
ATPase and acid secretion. digest proteins. Maximum peptic activity
In a study by Besancon et al” in porcine occurs at a pH of 1.5 to 2. Increasing the
gastric vesicles, rabeprazole fully inhibited pH thus decreases the activity of pepsin,
H+,K+-ATPase within 5 minutes, whereas which, because pepsin is a major compo-
omeprazole and lansoprazole required 30 nent of the refluxate in GERD, subse-
minutes to do so, and pantoprazole had quently decreases esophageal erosion.’
reached only 50% inhibition at 4.5 minutes. GERD symptoms such as heartburn are
Inhibition of acid secretion correlated with also associated with more acidic reflux-
rabeprazole’s ability to bind to a cysteine ate. Patients with duodenal ulcer gener-
residue (either cysteine-8 13 or cysteine- ally secrete more gastric acid than healthy
822) in the fifth to sixth transmembrane subjects, but there is much interpatient
segment of the alpha domain of H+,K+- variability. Ulcer pain seems to correlate
ATPase. Even after full inhibition of acid with acid secretion in most patients. Pa-
secretion had occurred, rabeprazole con- tients with gastric ulcer usually have nor-
tinued to bind to other exposed cysteines mal or low levels of acid secretion, but
in the enzyme: cysteine-892 in the seventh gastric acid can contribute to gastric ul-
to eighth transmembrane segment and cys- cers because it disrupts mucosal integrity.6
teine-321 in the third transmembrane seg- The presence of Helicobacter pylori is
ment. In comparison, although other PPIs now thought to cause 270% of gastric ul-
also interacted initially with 1 of the 2 cys- cers and 90% of duodenal ulcers. Pres-
teines in the fifth to sixth transmembrane ence of this microorganism increases the
segment to inhibit acid secretion, panto- secretion of gastric acid and also affects
prazole did not bind to other exposed cys- the antisecretory activity of PPIs. There-
teines, whereas omeprazole also interacted fore, achieving adequate acid suppression
with cysteine-892 and lansoprazole with to produce ulcer healing is pharmacolog-
cysteine-32 1. The pharmacologic signifi- ically complex. The relationship of gas-

268
J. HORN

tric acid suppression to gastric ulcer heal- with rabeprazole than omeprazole (P <
ing is particularly complex, with healing 0.001) (Figure 2). After 8 days, rabepra-
usually requiring a longer period of sup- zole continued to produce an intragastric
pression in patients with gastric ulcer than pH ~3 and >4 for a greater percentage of
in those with duodenal ulcer.6 time. Although median 24-hour intragas-
tric pH did not differ significantly between
treatments, rabeprazole provided a signif-
Proton-Pump Inhibitors and Gastric
icantly greater decrease in acidity during
Acid Secretion
the afternoon and nighttime periods. This
As a class, PPIs are more effective than study demonstrated that rabeprazole 20
histamine, (HJ-receptor antagonists in el- mg had a significantly faster onset of acid
evating 24-hour intragastric pH, but indi- control than did omeprazole 20 mg.
vidual agents differ in time to onset of acid The results of other studies9*‘0 suggest
control. In a placebo-controlled, crossover that pantoprazole and lansoprazole may
study, Williams et al8 measured 24-hour be more effective than omeprazole in el-
intragastric acidity after 1 and 8 days of evating 24-hour intragastric pH after 7
once-daily treatment with rabeprazole 20 days of treatment. However, these differ-
mg or omeprazole 20 mg in a group of 24 ences may reflect the doses compared.
healthy men. After the first dose, the me- For example, pantoprazole 40 mg and
dian 24-hour intragastric pH and percent- lansoprazole 30 mg were more effective
age of time during which intragastric pH in inhibiting acid secretion in healthy
was ~3 and >4 were significantly greater volunteers than omeprazole 20 mg, but

5 n Rabeprazole 20 mg
n Omeprazole 20 mg

--
Day 1 Day 8

Figure 2. Median 24-hour intragastric pH after 1 and 8 days of rabeprazole 20 mg,

omeprazole 20 mg, or placebo in healthy volunteers.8 *P < 0.001 versus placebo;
'P < 0.001 versus omeprazole.

269
 CLINICAL THERAPEUTICY

were equally as effective as omeprazole GERD is related to the duration of acid sup-
40 mg.9g10 pression within a 24-hour period. Robinson
As mentioned, the presence of H pylori et alI6 found that administration of rabepra-
appears to affect the antisecretory activity zole 20 or 40 mg once daily for 7 days in
of PPIs. In studies of omeprazole and pan- patients with GERD increased mean 24-
toprazole treatment, intragastric pH was hour intragastric pH to 4.2 and 4.7, respec-
significantly higher in subjects with Hpy- tively. These increases in intragastric pH
lori infection than in other subjects (P < were accompanied by reductions in the per-
0.01). In the omeprazole study, for exam- centage of time that esophageal pH was ~4,
ple, the median daytime pH in the corpus as well as in the overall number of reflux
was 5.4 in H pylori-positive subjects and episodes and prolonged reflux episodes.
2.9 in H pylori-negative ones.” H py- In the majority of patients, healing oc-
lori-related differences in intragastric pH curs after 4 weeks of once-daily treatment
were evident at both low and high doses with a PPI, compared with 8 weeks of
of pantoprazole.t2 treatment several times daily with an H,-
In a study in ulcer patientsI intragastric receptor antagonist.’ Comparative clinical
pH following administration of omeprazole trials show PPIs to be at least as effective
was higher before eradication of H pylori as Hz-receptor antagonists, if not more so,
than after eradication. However, in the ab- and they are equally well tolerated. In a
sence of omeprazole, intragastric pH did not multicenter study involving 202 patients
differ before or after eradication of H py- with erosive or ulcerative GERD, Dekkers
lon’. Possible explanations for the apparent et alI7 found that the healing rates for
greater effect of PPIs in H pylori-positive rabeprazole 20 mg and omeprazole 20 mg
subjects are that the bacterium produces acid- were equivalent in patients with GERD:
neutralizing substances such as ammonia 8 1% after 4 weeks of treatment and >90%
that may also interfere with H+,K+-ATPase, after 8 weeks of treatment (Figure 3). Both
and that the gastritis associated with the in- PPIs also provided similar relief of heart-
fection promotes release of acid-inhibiting burn, as measured by frequency and sever-
cytokines.14 In any case, PPI doses may ity of symptoms. In another multicenter
need to be higher in patients who do not study, Mijssner et alI8 found that the
harbor H pylori than in those who do. length of time required for symptom re-
lief was similar for pantoprazole 40 mg
and omeprazole 20 mg. The 2 treatments
CLINICAL INDICATIONS FOR
provided similar healing rates of 74% and
PROTON-PUMP INHIBITORS
78%, respectively, after 4 weeks of treat-
ment, and 90% and 94% after 8 weeks.
Gastroesophageal Rejlux Disease
GERD healing is maintained and symp-
GERD, which is characterized by pro- toms are prevented in the majority of pa-
longed and repeated exposure of the tients who undergo long-term treatment
esophageal mucosa to acidic gastric con- with PPIs. In patients who received 1 year
tents, is often accompanied by erosive or of daily treatment for GERD and had en-
ulcerative damage to the esophagus. In a doscopic evidence of healing, the relapse
meta-analysis of numerous clinical trials, rate was significantly lower with lansopra-
Bell et alI5 found that esophageal healing in zole or omeprazole than with placebo or ra-

270
J. HORN

W Rabeprazole 20 mg
q Omeprazole 20 mg

GERD Gastric Ulcer Duodenal Ulcer

98

O-
4 8 3 2 4
Time (wky

Figure 3. Comparison of healing rates after once-daily treatment with rabeprazole 20 mg

and omeprazole 20 mg in 202 patients with ulcerative or erosive gastro-
esophageal reflux disease (GERD), I7 227 patients with active gastric ulcer,22
and 205 patients with active duodenal ulcer.2s

nitidine.‘O Rabeprazole has also been shown correlation between suppression of 24-
to be highly effective in long-term preven- hour intragastric acidity and the rate of
tion of esophagitis and GERD symptoms.‘” gastric ulcer healing after 2 (P = O.Ol), 4
In various comparative triaI~,‘~ the relapse (P < O.Ol), and 8 weeks (P < 0.01) of
rates for once-daily lansoprazole 30 mg and treatment. The pharmacologic agents
omeprazole 20 mg were similar, and those studied, which included omeprazole, var-
for lansoprazole were significantly lower ious H2-receptor antagonists, and enpros-
than those for twice-daily ranitidine 300 til, suppressed 24-hour intragastric acidity
mg (P < 0.01). These studies demonstrate by 34% to 98%. This acid suppression
the effectiveness of PPI therapy in main- was associated with ulcer healing rates of
taining GERD healing. 21% to 44% at 2 weeks, 50% to 80% at 4
weeks, and 82% to 96% at 8 weeks. Thus,
the rate of ulcer healing was related to the
Gastric Ulcer
duration of treatment.
In a meta-analysis of studies involving PPIs provide more rapid healing and
a variety of antisecretory drug regimens, higher ulcer healing rates than H,-recep-
Howden and Hunt20 found a significant tor antagonists. Studies have demon-

271
 CLINICAL THERAPEUTICS@

strated that once-daily treatment with clinical studies, once-daily rabeprazole 20

omeprazole, lansoprazole, or pantopra- mg was more effective than twice-daily
zole is superior to twice-daily treatment ranitidine 150 mg in patients with duode-
with ranitidine, famotidine, or cimeti- nal ulcer.24 After 2 and 4 weeks of treat-
dine.21 In most comparative studies in pa- ment, once-daily pantoprazole 40 mg was
tients with active gastric ulcer, individual more effective than once-daily ranitidine
PPIs have exhibited similar rates of ulcer 300 mg in producing duodenal ulcer heal-
healing. For example, in a randomized, ing, and many (but not all) studies showed
double-blind, multicenter study involving greater pain relief with pantoprazole.’
227 patients with active gastric ulcers, Moreover, 8 weeks of PPI treatment pro-
Dekkers et alz2 found that once-daily treat- duces ulcer healing in the majority of pa-
ment with either rabeprazole 20 mg or tients who are refractory to treatment with
omeprazole 20 mg produced equivalent H,-receptor antagonists. lo
endoscopic healing rates after 3 weeks In a randomized, double-blind, multi-
(58% vs 61%) and 6 weeks (91% each) center study, Dekkers et a12” compared
(see Figure 3). However, rabeprazole pro- rabeprazole 20 mg and omeprazole 20 mg
vided more consistent, and in some cases in 205 patients with active duodenal ulcer,
statistically superior, improvement in 82% of whom had H pylori infection. Re-
symptoms compared with omeprazole at spective ulcer healing rates at 2 and 4
both time points. Approximately half of weeks were 69% and 98% with rabepra-
the patients in this study had antibodies zole and 61% and 93% with omeprazole
against H pylori. (see Figure 3). By the end of the study, pa-
tients treated with rabeprazole had a sig-
nificantly greater improvement in daytime
Duodenal Ulcer
pain compared with those given omepra-
Treatment of patients with duodenal ul- zole (P = 0.04). In studies comparing pan-
cer requires eradication of H pylori to re- toprazole 40 mg and omeprazole 20 mg,
duce the risk of relapse. However, the ma- the 2 PPIs provided similar rates of ulcer
jority of comparative clinical trials have healing and symptom relief.9
not addressed the role of H pylori. One
meta-analysis of a large number of clini-
Zollinger-Ellison Syndrome
cal studies23 found that elevation of intra-
gastric pH to >3 for 218 hours each day Sustained release of gastrin from gas-
was optimal for duodenal ulcer healing. trinomas in the pancreas or duodenum is
However, as will be described, eradica- central to the pathogenesis of Zollinger-
tion of H pylori with combination antibi- Ellison syndrome. Gastrin increases basal
otic therapy may require elevation of in- acid secretion, causing hypertrophy of the
tragastric pH to ~5.’ gastric mucosa, which in turn causes an
Duodenal ulcer healing is achieved af- increase in the number of parietal cells
ter 2 to 4 weeks of treatment with once- and therefore increased maximum acid
daily PPI therapy, compared with 4 to 8 output. Unregulated hypersecretion of
weeks of treatment with H,-receptor an- acid and pepsin leads to severe peptic ul-
tagonists.’ Moreover, symptom relief oc- cer disease, esophagitis, and diarrhea. Be-
curs more rapidly with PPI therapy. In cause gastrinomas tend to grow slowly,

272
J. HORN

the early morbidity and mortality of this previous gastrectomy or esophagitis. Lan-
syndrome are attributable to acid hyper- soprazole produced ulcer healing and
secretion and ulcer complications.26-28 symptom resolution, usually within sev-
Treatment goals are twofold: control of eral weeks of beginning treatment. No-
acid hypersecretion, and location and pos- tably, patients with Zollinger-Ellison syn-
sible removal of the gastrinoma.27 How- drome who had undergone gastrectomy
ever, the 5-year cure rate can be <30% af- were more likely to experience ulcer re-
ter gastrinoma resection, even among currence, even with strict acid control.
patients who have no metastases.26There- Rabeprazole was administered to 10 pa-
fore, control of acid to achieve symptom tients with Zollinger-Ellison syndrome or
relief and ulcer healing is central to the idiopathic gastric acid hypersecretion at
management of this disease. Acid secre- an initial dose of 60 mg daily (or 40 mg
tion must be reduced to ~10 mmol/h to twice daily in patients who had had pre-
prevent complications.28 Although the in- vious gastric surgery).3’ The dose was in-
troduction of Hz-receptor antagonists was creased as necessary to maintain acid se-
a major treatment advance, the necessary cretion at Cl0 mmol/h (or <5 mmol/h) for
level of acid suppression was difficult to up to 12 months. Only 2 patients required
achieve, even with dosing 4 times daily. a daily dose >60 mg, and no patients had
Oral Hz-receptor antagonists are no longer a recurrence of peptic acid disease over
used to treat patients with Zollinger-Ellison the course of the study.
syndrome, because PPIs are better able to
suppress acid secretion and have a longer
Barrett’s Esophagus
duration of action. With PPIs, it is gener-
ally possible to reduce basal acid secre- Barrett’s esophagus is characterized by
tion to 12 mrr~ol/h.~~ the presence of columnar epithelium in
Long-term treatment with omeprazole, the lower esophagus instead of the normal
lansoprazole, or rabeprazole is effective in squamous lining. The condition is associ-
the management of this disease. The results ated with gastroesophageal reflux and an
of several studies involving 210 patients increased risk of esophageal cancer. The
with Zollinger-Ellison syndrome showed disorder cannot be distinguished from
omeprazole to be effective in 99% of pa- GERD based on symptoms alone; en-
tients over a period ranging from 0.5 to 54 doscopy and biopsy are needed to con-
montbs.29 In these studies, omeprazole was firm the diagnosis. Nevertheless, the treat-
administered at median daily doses of 60 to ment of acid reflux is the same in either
100 mg, with 20% to 60% of patients re- case, with PPIs being more effective than
ceiving treatment in divided daily doses. H,-receptor antagonists administered
In another study, 3olansoprazole was ad- alone or in combination with cisapride.
ministered for a median of 28 months to PPIs are also effective in many patients in
26 patients with Zollinger-Ellison syn- whom the condition has been refractory
drome and peptic ulcer disease. The ini- to treatment with Hz-receptor antagonists.
tial daily dose of 60 mg was individual- Symptoms disappear with treatment in
ized to achieve a basal acid output of <5 most patients but may recur within days
mmol/h in patients with intact stomachs, of discontinuing therapy. Therefore, long-
and ~1 mmol/h in patients who had had a term use of PPIs may be required.32,33

273
 CLINICAL THERAPEUTICS@

Although there is some recent evidence omeprazole and pantoprazole, respec-

that continuous PPI therapy for several tively.“* In a smaller study,39 rabeprazole
years can cause partial regression of Bar- was 2 times more active than lansopra-
rett’s esophagus,34,35 this effect seems to zole and 4 times more active than omepra-
be modest, with the overall length of the zole. The growth-inhibitory effect of PPIs
columnar epithelium not changing dra- may be sufficient to enable eradication of
matically. Surgery, an alternative treat- H pylori by coadministered antibiotics.
ment approach, is the same as for GERD PPIs may inhibit H pylori growth by
unless severe dysplasia occurs, in which several mechanisms. H pylori produces a
case esophagectomy may be indicated. urease that catalyzes the hydrolysis of urea
Overall, surgical treatment might provide to ammonia, which is believed to protect
better results than long-term medical treat- the bacterium against acid and allow it to
ment with antisecretory agents in terms of colonize the gastric mucosa. PPIs potently
controlling disease activity, esophagitis, inhibit H pylori urease activity, presum-
and acid reflux.“” ably by binding to the active site of the
enzyme. Rabeprazole, with an IC,, (con-
centration required to produce 50% inhi-
ERADICATION OF
bition of urease) of 0.29 pmol/L, was - 10
HELICOBACTER PYLORI
times more potent than omeprazole and
H pylori is strongly associated with peptic lansoprazole in inhibiting urease at pH 5
ulcer disease and appears to play a signif- in both cellular and cell-free systems in
icant role in its pathogenesis. Eradication vitro.40 However, omeprazole was also
of this organism is strongly recommended shown to inhibit H p-ylori via a urease-
in the management of gastric or duodenal independent mechanism, with inhibition
ulcer, whether initial or recurrent.36,37 An- of growth seen at a low pH both in the ab-
tibiotic monotherapy, however, provides sence of urea and in a urease-deficient
poor eradication rates, despite strong in strain of H pylori.4’
vitro activity against H pylori, and con- A second target of PPIs in H pylori may
comitant administration of multiple antibi- be a membrane-bound H+,K+-ATPase,
otics causes unacceptable side effects.33,38 similar to that found in human parietal
Coadministration of antibiotics with a PPI, cells, that may help the bacterium main-
discussed later, appears to have a syner- tain a large proton gradient in an acidic
gistic effect and yields high Hpylori erad- environment. PPIs have been found to
ication rates. strongly inhibit this ATPase at a pH of 4,
although they were essentially inactive at
a neutral pH.42
Antibacterial Effects
PPIs inhibit the growth of H pylori at
Proton-Pump Inhibitor-Based
concentrations similar to those that inhibit
Combination Therapy
gastric H+,K+-ATPase. Against a panel of
58 clinical isolates of H pylori, lansopra- Eradication of H pylori with PPI-based
zole inhibited 90% of organisms at a min- dual-, triple-, and quadruple-drug regi-
imum concentration of 6.25 p,g/mL, and mens has been evaluated in many ran-
was 4 and 16 times more active than domized clinical studies. The results of

274
J. HORN

several recent studies43-49are summarized continued for several additional weeks in

in the table. According to the Maastricht some of the studies of ulcer healing.46,48
Consensus Report,37 the recommended Antibiotic therapy differed in terms of
regimen for H pylori eradication is PPI- drugs used, dose, number of times admin-
based triple-drug therapy for 7 days, con- istered daily, and duration of treatment.
sisting of a PPI plus 2 of the following In a review of 275 treatment arms in-
antibiotics: clarithromycin, amoxicillin, volving omeprazole-based therapy,“O
or a nitroimidazole (ie, metronidazole or triple-drug therapy produced median erad-
tinidazole). ication rates of 86%, and dual-drug ther-
Stack et a143 evaluated rabeprazole- apy produced median eradication rates of
based dual- and triple-drug therapy in 75 58%. Quadruple-drug therapy was no
H pylori-infected patients with chronic more effective than triple-drug therapy.
gastritis. Patients were randomized to 7 Median eradication rates were similar
days of twice-daily treatment with 1 of 4 with treatment for 1 and 2 weeks; were
antibiotic regimens consisting of rabepra- slightly higher with a daily dose of 40 mg
zole 20 mg plus: (1) clarithromycin 500 compared with 20 mg; and were compa-
mg (RC); (2) amoxicillin 1 g and clarith- rable with regimens including 2 of the fol-
romycin 500 mg (RAC); (3) amoxicillin lowing: clarithromycin, amoxicillin, and
1 g and metronidazole 400 mg (RAM); or metronidazole (or tinidazole). Taken to-
(4) clarithromycin 500 mg and metronida- gether, these results are consistent with
zole 400 mg (RCM). Intent-to-treat analy- those in the Maastricht Consensus Re-
sis showed eradication rates of 100% with port37: PPI-based triple-drug therapy for 7
RCM, 95% with RAC, and 90% with days is effective in eradicating H pylori.
RAM. The dual-drug treatment, RC, was
less effective (63%), and resistance to cla-
DRUG INTERACTIONS
rithromycin developed in the majority of
patients in whom this treatment failed. The Drug-drug interactions vary among the
patients for whom RAM failed were in- PPIs. Omeprazole has been shown to in-
fected with metronidazole-resistant strains teract with diazepam, phenytoin, and war-
at the inception of treatment. Thus, rabepra- farin, but not with theophylline. Lansopra-
zole-based triple-drug therapy for 7 days zole causesa small decrease in theophylline
was effective in eradicating H pylori. concentration and may decrease the effi-
In a number of recent studies,43-50triple- cacy of oral contraceptives. Pantoprazole
drug therapy with other PPIs has been and rabeprazole appear to be free of these
more effective in eradicating Hpylori than interactions. The clinical significance of
dual-drug therapy. In general, such ther- these interactions is not yet known, but
apy produced eradication rates of 279%, careful monitoring is recommended.51,52
whereas dual-drug regimens produced
eradication rates of 53% to 77% (see
DISCUSSION AND CONCLUSIONS
table). However, it is difficult to compare
the various PPI-based regimens, because The various PPIs have much in common,
these studies differed in several important notably their superiority to Hz-receptor an-
ways. The duration of treatment ranged tagonists in the treatment of acid-related
from 7 to 14 days, but PPI therapy was diseases. In GERD and peptic ulcer dis-

275
Table. Recent studies of proton-pump inhibitor (PPI)-based antibiotic therapy for the eradication of Helicobacter pylori (intent-
to-treat analysis).

Antibiotic
Duration Eradication
PPI Regimen Regimen (4 Rate (%) Reference

Rabeprazole 20 mg BID C 500 mg BID + M 400 mg BID 7 100 Stack et al43

A 1 g BID + C 500 mg BID 7 95
A 1 g BID + M 400 mg BID 7 90
C 500 mg BID 7 63
Omeprazole 20 mg BID C 500 mg BID + M 400 mg BID 7 92 Chu et a144
A 1 g BID 14 74
Lansoprazole 30 mg QD* B 120 mg QID + M 400 mg BID +
TC 500 mg QID 7 90 Korman et a149
Lansoprazole 30 mg BID A 1 g BID + C 500 mg BID 14 94 Schwartz et a145
C 500 mg BID 14 57
C 500 mg TID 14 7.5
IA I gTID 14 53
Lansoprazole 30 mg TID A I g TID 14 77 Schwartz et al45
P
None 14 2 P
Pantoprazole 40 mg BID A 1 g BID + C 500 mg BID 14 91 Louw et al47 E
A I g BID + C 500 mg BID 7 79 c
Pantoprazole 40 mg BID C 500 mg TID + M 500 mg TID 7 95 Adamek et a146 2
C 500 mg TID 14 60 6
Pantoprazole 40 mg QD C 250 mg BID + T 500 mg BID 7 86 Pazzi et a14X %
2
C = clarithromycin; M = metronidazole; A = amoxicillin; B = bismuth subcitrate; TC = tetracycline; T = tinidazole.
*Lansoprazole was administered for IO days, beginning 3 days before antibiotics were started.
J. HORN

ease, once-daily administration of PPIs is The in vitro potency and binding patterns
faster and more effective than treatment of lansoprazole are similar to those of
with Hz-receptor antagonists several times omeprazole. Like the other PPIs, lansopra-
daily. In clinical trials, PPIs administered zole is effective in treating GERD and pep-
once daily produced endoscopic evidence tic ulcer. Both lansoprazole and omeprazole
of healing in >90% of patients with duo- have been shown to be effective in the long-
denal ulcer after 4 weeks of treatment, term treatment of Zollinger-Ellison syn-
>90% of those with gastric ulcer after 6 drome. In vitro studies indicate that lanso-
weeks of treatment, and >90% of those prazole is more active against H pylon’ than
with ulcerative or erosive GERD after 8 omeprazole and pantoprazole, although not
weeks of treatment. Maintenance therapy as active as rabeprazole.
with daily PPIs is effective in preventing In GERD and peptic ulcer, pantopra-
GERD relapse. Comparative clinical tri- zole appears to provide symptom relief
als indicate that individual PPIs produce and healing similar to those produced by
equivalent ulcer healing, but it is impor- omeprazole. The in vitro activity of pan-
tant to recognize that these studies were toprazole is unusual, given its slower cys-
designed to demonstrate equivalence. teine-binding pattern, but additional stud-
Although they share a common struc- ies are needed to ascertain the significance
ture and mode of action, the PPIs differ of this characteristic.
in their clinical pharmacology. Rabepra- Omeprazole is the best-established and
zole is a more rapid and potent inhibitor therefore most-studied drug in the PPI class.
of H+,K+-ATPase than omeprazole, pos- Its effectiveness in the treatment of all acid-
sibly because of its faster activation in related diseases is well documented. Ide-
the parietal cell canaliculus. After the first ally, the differences between omeprazole
dose, rabeprazole produces greater inhi- and other PPIs should be used to provide
bition of acid secretion than omeprazole, the most advantageous treatment in indi-
as reflected in a higher 24-hour intragas- vidual cases, but further study of their phar-
tric pH and longer periods in which pH is macologic differences is required before
>3 and ~4.~ This more rapid onset of acid such optimal use can be realized.
control may be particularly relevant in
PPI-based triple-drug therapy for H py-
ACKNOWLEDGMENT
lori eradication when a short 7-day treat-
ment course is recommended. One such The author is a member of the speakers’bu-
regimen, rabeprazole plus clarithromycin reau of Eisai-Janssen,Teaneck, New Jersey,
and metronidazole, was found to eradi- and has served as a consultant to Janssen
cate H pylori in all patients treated.4” Pharmaceutics, Titusville, New Jersey.
Clinical trials comparing rabeprazole-
based triple-drug therapy with other PPI-
based triple-drug therapies are needed to Address correspondence to: John Horn,
ascertain whether rabeprazole’s more PharmD, Department of Pharmacy,
rapid onset of acid control actually does University of Washington School of
provide a higher Hpylori eradication rate, Pharmacy, Health Sciences Building,
and to define the general clinical rele- Room T-341, 1959 NE Pacific Street,
vance of this more rapid onset. Seattle. WA 98 185.

277
 CLINICAL THERAPEUTICS@

REFERENCES and clinical efficacy in the management

of acid-related disorders. Drugs. 1997;54:
1. Sachs G. Proton pump inhibitors and acid- 473-500.
related diseases. Pharmacotherupy. 1997;
17:22-37. 11. Verdh EF, Armstrong D, Fraser R, et al.
Effect of Helicohacter pylori status on in-
2. Sachs G, Shin JM, Briving C, et al. The
tragastric pH during treatment with
pharmacology of the gastric acid pump:
omeprazole. Gut. 1995;36:539-543.
The H+,K+ ATPase. Annu Rev Pharmncol
Toxicol. 1995;35:277-305. 12. Koop H, Kuly S, Fhig M, et al. Intragas-
tric pH and serum gastrin during adminis-
3. Fujisaki H, Shibata H, Oketani K, et al.
Inhibitions of acid secretion by E38 10 and tration of different doses of pantoprazole
omeprazole, and their reversal by glu- in healthy subjects. Eur J Gastroenterol
tathione. Biochem Pharmacol. 199 1;42: Hepatol. 1996;8:915-918.
321-328.
13. Verdli EF, Armstrong D, Idstrom J-P, et al.
4. Williams MP, Pounder RE. The pharma- Effect of curing Helicobacter pylori infec-
cology of rabeprazole. Aliment Pharma- tion on intragastric pH during treatment
co1 Ther: 1999;13(Suppl 3):3-10. with omeprazole. Gut. 1995;37:743-748.

5. Besancon M, Simon A, Sachs G, Shin JM. 14. Smout AJPM. Is the sensitivity to gastric
Sites of reaction of the gastric H,K-ATPase acid inhibition Helicobacter pylori sta-
with extracytoplasmic thiol reagents. J Biol tus-dependent? Stand J Gastroenterol.
Chem. 1997;272:22,438-22,446. 1998;33(Suppl 225):32-35.

6. Huang JQ, Hunt RH. pH, healing rate and 15. Bell NV, Burget D, Howden CW, et al.
symptom relief in acid-related diseases. Appropriate acid suppression for the man-
Yale J Biol Med. 1996;69: 159-174. agement of gastro-oesophageal reflux dis-
ease. Digestion. 1992;51(Suppl 1):59-67.
7. Goldberg HI, Dodds WJ, Gee S, et al. Role
of acid and pepsin in acute experimental 16. Robinson M, Maton PN, Rodriguez S, et
esophagitis. Gastroenterology. 1969;56: al. Effects of oral rabeprazole on oe-
223-230. sophageal and gastric pH in patients with
gastro-oesophageal reflux disease. Ali-
8. Williams MP, Sercombe J, Hamilton MI,
ment Pharmacol Ther: 1997; I I :973-980.
Pounder RE. A placebo-controlled trial to
assess the effects of 8 days of dosing with
17. Dekkers CPM, Beker JA, Thjodleifsson
rabeprazole versus omeprazole on 24-h
B, et al. Double-blind, placebo-controlled
intragastric acidity and plasma gastrin
comparison of rabeprazole 20 mg vs.
concentrations in young healthy male sub-
omeprazole 20 mg in the treatment of ero-
jects. Aliment Pharmacol Ther: 1998; 12:
sive or ulcerative gastro-oesophageal re-
1079-1089.
flux disease. Aliment Pharmacol Ther.
9. Fitton A, Wiseman L. Pantoprazole. A re- 1999;13:49-57.
view of its pharmacological properties and
therapeutic use in acid-related disorders. 18. Mossner J, Holscher AH, Herz R, Schnei-
Drugs. 1996;5 1:460-482. der A. A double-blind study of pantopra-
zole and omeprazole in the treatment of
10. Langtry HD, Wilde MI. Lansoprazole: An reflux oesophagitis: A multicentre trial. Al-
update of its pharmacological properties iment Pharmacol Ther: 1995;9:321-326.

278
J. HORN

19. Birbara C, Breiter J, Collins D, et al. this complex disorder. Postgrad Med.
Rabeprazole: Preventing endoscopic and 1998;104:155-158,163-164.
symptomatic relapse in erosive or ulcerative
GERD. Am J Gastroenterol. 199893: 1630. 28. Maton PN. Zollinger-Ellison syndrome:
Abstract. Recognition and management of acid hy-
persecretion. Drugs. 1996;52:3344.
20. Howden CW, Hunt RH. The relationship
between suppression of acidity and gas- 29. Frucht H, Maton PN, Jensen RT. Use of
tric ulcer healing rates. Aliment Pharma- omeprazole in patients with Zollinger-
co1 Ther: 1990;4:25-33. Ellison syndrome. Dig Dis Sci. 1991;36:
394-404.
21. Richardson P, Hawkey CJ, Stack WA. Pro-
ton pump inhibitors: Pharmacology and
30. Hirschowitz BI, Mohnen J, Shaw S. Long-
rationale for use in gastrointestinal disor-
term treatment with lansoprazole for pa-
ders. Drugs. 1998;56:307-335.
tients with Zollinger-Ellison syndrome. Al-
22. Dekkers CPM, Beker JA, Thjodleifsson iment Pharmacol Ther 1996;10:507-522.
B, et al. Comparison of rabeprazole 20
mg vs. omeprazole 20 mg in the treatment 31 Mignon M, Merrouche M, Gardner J, et
of active gastric ulcer: A European multi- al. Rabeprazole therapy effective in
centre study. Aliment Pharmacol Thel: Zollinger-Ellison syndrome and idiopathic
1998;12:789-795. gastric acid hypersecretion. Gastroen-
terology. 1999; 116:A253. Abstract.
23. Burget DW, Chiverton SG, Hunt RH. Is
there an optimal degree of acid suppres- 32 Cameron AJ. Management of Barrett’s
sion for healing of duodenal ulcers? A esophagus. Mayo Clin Proc. 1998;73:
model of the relationship between ulcer 457461.
healing and acid suppression. Gastroen-
terology. 1990;99:345-35 1. 33. Bremner CG, Bremner RM. Barrett’s
esophagus. Surg Clin North Am. 1997;77:
24. Humphries TJ, Spera A, Breiter J, et al. 1115-1137.
Rabeprazole sodium once daily is supe-
rior to ranitidine 150mg bid in the healing 34. Peters FIM, Ganesh S, Kuipers EJ, et al.
of active duodenal ulcer. Gastroenterol- Endoscopic regression of Barrett’s esoph-
ogy. 1997;112:A154. Abstract. agus during omeprazole treatment: A ran-
domized double blind study. Gut. 1999;
25. Dekkers CPM, Beker JA, Thjodleifsson
45:489-494.
B, et al. Comparison of rabeprazole 20
mg versus omeprazole 20 mg in the treat-
35. Wilkinson SP, Biddlestone L, Gore S,
ment of active duodenal ulcer: A Euro-
Shepherd NA. Regression of columnar-
pean multicentre study. Aliment Pharma-
lined (Barrett’s) oesophagus with omepra-
co1 Thel: 1999;13:179-186.
zole 40 mg daily: Results of 5 years of
26. Hirschowitz BI. Zollinger-Ellison syn- continuous therapy. Aliment Pharmacol
drome: Pathogenesis, diagnosis, and man- Thel: 1999;13:1205-1209.
agement. Am J Gastroenterol. 1997;92
(Suppl 4):44S-50s. 36. NIH Consensus Development Panel on
Helicobacter pylori in Peptic Ulcer Dis-
21. Qureshi W, Rashid S. Zollinger-Ellison ease. Helicobacter pylori in peptic ulcer
syndrome: Improved treatment options for disease. JAMA. 1994;272:65-69.

279
 CLINICAL THERAPEUTICS”

37. The European Helicobacter pylori Study Helicobacter pylon’ and preventing ulcer
Group. Current European concepts in the recurrence: A randomized, double-blind,
management of Helicobacter pylori infec- multicenter study of lansoprazole, clarith-
tion: The Maastricht Consensus Report. romycin, and/or amoxicillin in different
Gut. 1997;41:8-13. dosing regimens. Am J Gastroenterol.
1998;93:584-590.
38. Nakao M, Malfertheiner P. Growth in-
hibitory and bactericidal activities of lan- 46. Adamek RJ, Bethke TD, and the Intema-
soprazole compared with those of omepra- tional Pantoprazole HP Study Group. Cure
zole and pantoprazole against Helicobacter of Helicobacter pylori infection and heal-
pylori. Helicobacter: 1998;3:2 l-27. ing of duodenal ulcer: Comparison of pan-
toprazole-based one-week modified triple
39. Hirai M, Azuma T, Ito S, et al. A proton
therapy versus two-week dual therapy. Am
pump inhibitor, E3810, has antibacterial
J Gastroenterol. 1998;93: I9 19-l 924.
activity through binding to Helicobacter
pylori. J Gastroenterol. 1995;30:46 la64.
47. Louw JA, Van Rensburg CJ, Hanslo D, et
40. Tsuchiya M, Imamura L, Park J-B, Kobashi al. Two-week course of pantoprazole com-
K. Helicobacter pylori urease inhibition by bined with 1 week of amoxycillin and cla-
rabeprazole, a proton pump inhibitor. Biol rithromycin is effective in Helicobacter py-
Pharm Bull. 1995;18:1053-1056. lori eradication and duodenal ulcer healing.
Aliment Pharmacol Ther: 1998; 12:545-550.
41. McGowan CC, Cover TL, Blaser MJ. The
proton pump inhibitor omeprazole inhibits 48. Pazzi P, Scagliarini R, Gamberini S, et al.
acid survival of Helicobacter pylori by a Short-term, low-dose pantoprazole-based
urease-independent mechanism. Gastro- triple therapy for cure of Helicobacter py-
enterology. 1994; 107:738-743. lon’ infection in duodenal ulcer patients. Al-
iment Pharmacol The,: 1998; 12:731-734.
42. Mauch F, Bode G, Malfertheiner P. Identi-
fication and characterization of an ATPase 49. Korman MG, Bolin TD, Nicholson FB,
system of Helicobacter pylori and the ef- Engelman JL. Lansoprazole quadruple
fect of proton pump inhibitors. Am J Gas- therapy is effective in curing Helicobacter
troenterol. 1993;88: 1801-l 802. Letter. pylori. Helicobactel: 1998;3:202-20.5.
43. Stack WA, Knifton A, Thirlwell D, et al.
50. Langtry HD, Wilde MI. Omeprazole: A
Safety and efficacy of rabeprazole in com-
review of its use in Helicobacter pylori
bination with four antibiotic regimens for
infection, gastro-oesophageal reflux dis-
the eradication of Helicobacter pylori in
ease and peptic ulcers induced by non-
patients with chronic gastritis with or
steroidal anti-inflammatory drugs. Drugs.
without peptic ulceration. Am J Gastroen-
I998;56:447486.
terol. 1998;93: 1909-1913.
44. Chu K-M, Choi H-K, Tuen HH, et al. A 51. Mears JM, Kaplan B. Proton pump in-
prospective randomized trial comparing hibitors: New drugs and indications. Am
the use of omeprazole-based dual and Fam Phys. 1996;53:285-292.
triple therapy for eradication of Heli-
cobacter pylori. Am J Gastroenterol. 52. Humphries T. A review of the drug-drug
1998;93:1436-1442. interaction potential of rabeprazole sodium
based on CYP-450 interference or absorp-
45. Schwartz H, Krause R, Sahba B, et al. tion effects. Digestion. 1998;59(Suppl):76.
Triple versus dual therapy for eradicating Abstract.

280