SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Aliment Pharmacol Ther 2010; 32: 1091–1101

Summary

Background  Nonsteroidal anti-inflammatory drugs are associated with upper gastrointestinal mucosal injury. Naproxen etemesil is a lipophilic, non-acidic, inactive prodrug of naproxen that is hydrolysed to pharmacologically active naproxen once absorbed. We hypothesized that with lesser topical exposure to naproxen from the prodrug, there would be reduced gastroduodenal mucosal injury compared with naproxen.

Aim  To compare the degree of endoscopic mucosal damage of naproxen etemesil vs. naproxen.

Methods  This multicentre, randomized, double-blind, double-dummy trial compared oral naproxen etemesil 1200 mg twice daily (= 61) with naproxen 500 mg twice daily (= 59) for 7.5 days in 120 healthy subjects (45–70 years; mean 51 years; 58% female) with baseline total modified gastroduodenal Lanza score ≤2 (no erosions/ulcers) on endoscopy. The primary endpoint was mean total modified gastroduodenal Lanza score on day 7. A secondary endpoint was incidence of gastric ulcers.

Results  The day 7 mean total modified gastroduodenal Lanza score was 2.8 ± 1.7 for naproxen etemesil vs. 3.5 ± 2.0 for naproxen (= 0.03), and significantly fewer naproxen etemesil-treated subjects (3.3%) developed gastric ulcers compared with naproxen-treated subjects (15.8%) (= 0.02).

Conclusion  In this first proof-of-concept study, naproxen etemesil was associated with significantly lower gastroduodenal mucosal injury compared with naproxen after 7 days of exposure (Clinical trial number: NCT00750243).


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Nonsteroidal anti-inflammatory drugs (NSAIDs) are one of the most frequently prescribed medications worldwide.1 Exposure to NSAIDs, however, is associated with an increased risk of endoscopic gastroduodenal mucosal injury and the development of serious upper gastrointestinal (UGI) events.2–6 Prospective studies have reproducibly demonstrated NSAID injury, with 20–30% of subjects who continuously use non-selective NSAIDs developing endoscopically proven ulcers.6 Epidemiological evidence has shown that NSAID-users have a 4-fold greater risk of UGI complications compared with non-users, including bleeding and perforation.7–9

Several potential underlying mechanism(s) have been forwarded to explain the development of NSAID-related GI mucosal injury.10–14 An important dogma has traditionally highlighted the systemic effects of inhibition of mucosal cyclooxygenase (COX)-1 mediated prostaglandin synthesis as a primary factor predisposing to NSAID-associated gastromucosal injury.12–15 At the level of the mucosa, the prostaglandin-facilitated mucus-bicarbonate-phospholipid barrier serves a protective function against intraluminal acid and other barrier-penetrating substances. Cyclooxygenase-1 (COX-1) inhibition and associated lower levels of mucosal prostaglandins lead to subsequent reduction and disruption of the barrier, predisposing to and allowing for injury to occur.12, 16, 17

The propensity for NSAID-induced gastric mucosal injury is also dependent on topical injury, which in turn is related to the physicochemical properties of the drug itself.11, 18, 19 The acidic nature of the NSAID as well as its hydrophobicity/lipophilicity in part predict the degree to which an individual NSAID is likely to disrupt the integrity of the gastric mucosa;11, 19–21 the more acidic the agent, the greater the likelihood for topical GI injury.11 Also, drugs with low lipid solubility appear to lead to greater mucosal damage than more lipophilic agents.11

Together these proposed multifaceted mechanism(s) for injury serve as the biological basis for several therapeutic approaches to mitigate the UGI effects of NSAIDs. These include formulation strategies (enteric-coating or sustained-release products), COX-2 selective inhibitors, pro-drugs, and co-prescription/co-delivery of a NSAID with a gastroprotective agent (e.g. misoprostol or proton pump inhibitors).6, 15, 22–24

Naproxen etemesil is a pharmacologically inactive ester prodrug of naproxen. Whereas naproxen is both acidic (pKa 4.2) and hydrophilic,11 the naproxen etemesil prodrug is non-acidic (pKa 7.51) and lipophilic.25 Naproxen etemesil is stable within the acidic environment of the UGI tract, and is absorbed as an intact molecule. Once absorbed, it is rapidly hydrolysed to naproxen by circulating esterases25 and derives its anti-inflammatory properties from the released naproxen. Studies comparing naproxen etemesil with equimolar doses of naproxen in rats have shown equivalent efficacy in both the carrageenan paw inflammation and adjuvant arthritis models. Animal model studies have also shown that naproxen etemesil caused fewer gastric and intestinal mucosal lesions than equimolar doses of naproxen (Data on file, Logical Therapeutics, Inc.).

Considering the physical and chemical attributes of naproxen etemesil, coupled with the animal safety data, this proof-of-concept investigation in humans was designed to test the hypothesis that naproxen etemesil would cause less gastric mucosal injury than naproxen. This trial enrolled healthy volunteers aged 45–70 years, and based the primary endpoint on overall endoscopic damage (modified gastroduodenal Lanza score)26, 27 after 7 days of exposure.

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Study design

This was a phase I, multicentre, randomized, double-blind, double-dummy, proof-of-concept study. Subjects were enrolled at six centres (three each in the United States and the Czech Republic). After a 14-day NSAID washout, a pre-treatment esophagogastroduodenoscopy (EGD) was performed within 2 days prior to the first dose of study medication. Eligible subjects without any endoscopic evidence of ulcers or erosions (= 120) were then randomized in a 1:1 ratio, stratified by Helicobacter pylori antibody status, to either oral naproxen etemesil 1200 mg twice daily or naproxen 500 mg twice daily for 7.5 days (total of 15 doses) (Figure 1). Given our interest in short-term exposure and potential differences in topical injury, we modelled and powered this proof-of-concept trial in concordance with previously reported 1 week endoscopy studies.27, 28

image

Figure 1.  Study design.

Download figure to PowerPoint

A post-treatment EGD was performed after the day 7 morning dose (total of 13 doses). Subjects who were found to have developed an ulcer (mucosal break measuring ≥3 mm in maximum diameter, with depth) then stopped study medications, and treatment for the ulcer(s) in these subjects could be initiated at the discretion of the investigator. Subjects without an ulcer on the EGD continued the dosing and underwent pharmacokinetic testing after the day 8 morning dose.25 All subjects (i.e. those without ulcer and those with ulcer) were then followed up until day 15 (end-of-study) for safety. This study was performed on an ambulatory population. However, dosing was observed in the clinic on the mornings of days 1, 6, 7 and 8, and on the evening of day 1. The study was conducted in accordance with the principles of the Declaration of Helsinki. Independent ethics committees and institutional review boards of each participating study centre approved the study protocol, and all subjects provided signed written informed consent prior to the screening procedures. This trial was registered on Clinicaltrials.gov, registration number NCT00750243.

Subjects

The study enrolled healthy male or female volunteers between 45 and 70 years of age with a BMI of 18–30 kg/m2. Key exclusion criteria were NSAID allergy, any gastroduodenal erosion or ulcer on pre-treatment EGD, history of gastroduodenal ulcer or GI disease, inflammatory bowel disease, malabsorption, cholecystectomy, history of GI surgery, recent surgery or fracture, and weight loss >10% of current weight within 12 weeks prior to screening. Current stable smokers of ≤10 cigarettes/day or ≤2 cigar or pipes daily were eligible. Subjects were prohibited from using non-selective NSAIDs (including aspirin), COX-2–selective inhibitors, anticoagulants, platelet inhibitors, proton pump inhibitors, type 2 histamine blockers (H2 blockers), selective serotonin reuptake inhibitors, serotonin norepinephrine reuptake inhibitors, oral bisphosphonates and antacids for 14 days prior to the baseline EGD and during the study. Additionally, for pharmacokinetic consideration, subjects were prohibited from using corticosteroids (other than inhaled or topical), insulin, anti-convulsants, phenobarbital drugs, diuretics, angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers, anti-depressants, sucralfate, cholestyramine, fibrates, ezetimibe, other binding agents, gastrointestinal motility agents, probenecid, allopurinol, HMG CoA reductase inhibitors (statins), chemotherapeutics, immunomodulatory medications, cholinesterase inhibitors, phenothiazines, amiodarone, and narcotic or non-narcotic analgesics other than acetaminophen/paracetamol for 60 days prior to baseline EGD and during the study. Likewise, alcohol, high fat foods, spicy foods (including curry and chilli), grapefruit (juice or fruit), and methylxanthine-containing beverages and foods were also prohibited within 48 h prior to baseline endoscopy and during the study.

Masking and treatments

All study subjects and study personnel were blinded to study treatments, except for the honest broker pharmacists at a single central Czech Republic location or the site-specific pharmacists in the United States. These pharmacists performed the web-based randomization and dispensed the study treatments. All study treatments were ingested every 12 h for 7.5 days (i.e. twice daily on days 1–7, and one dose only on the morning of day 8). The selection of the naproxen etemesil dose was based on modelling the best likelihood of achieving the same steady-state plasma exposure (AUC0–12) to naproxen as delivered by naproxen (Naprosyn) 500 mg. On a molar basis, 730 mg of naproxen etemesil is equivalent to 500 mg of naproxen (Data on file, Logical Therapeutics, Inc.). Subjects randomly assigned to the naproxen etemesil treatment group received three tablets of naproxen etemesil 400 mg (for a total of 1200 mg) and one dummy naproxen tablet at each dose. Subjects randomly assigned to the naproxen treatment group received one tablet of naproxen (Naprosyn; Roche, Welwyn Garden City, UK) 500 mg and three dummy naproxen etemesil tablets at each dose.

All doses of study medication were taken 10–20 min following completion of a meal with 250 mL of water. No food was then permitted for a minimum of 4 h after dosing. On day 7 only, the morning dose of study medications was taken while fasting, and the subject did not eat until 1–4 h after the EGD was completed.

Endoscopy procedures

Prior to the pre-treatment and day 7 EGDs, subjects fasted for at least 8 h. The pre-treatment EGD was performed on day −2, −1 or 1. The day 7 EGD was performed 1–4 h following the morning dose of study medication on day 7. The EGD was performed using a nasal endoscope with local topical anaesthetic or oral entry endoscope with the option of using a short-acting benzodiazepine or narcotic agent. For each subject, the baseline and day 7 EGDs were performed by the same endoscopist, using the same type of endoscope. Based on endoscopic findings of petechiae, erosions, and ulcers, the results were tabulated using a modified Lanza scoring system (Table 1).26, 27 In this study, we used the standardized definition of an endoscopic ulcer, a mucosal break measuring ≥3 mm in its greatest diameter, with depth.29, 30

Table 1.   Modified Lanza score26, 27
Lanza scoreDefinition
  1. Petechia = intramucosal haemorrhage with no overlying mucosal break; erosion = mucosal break with no depth; ulcer = mucosal break measuring ≥3 mm in maximum diameter, with depth.

0No visible lesions
11–10 petechiae
2>10 petechiae
31–5 erosions
46–10 erosions
511–25 erosions
6>25 erosions
7Ulcer

For quality assurance purposes and to reduce inter-observer variability, each participating endoscopist was required to complete and document completion of a web-based training for use of the modified Lanza scoring system with JLG (author). A review of the endoscopist’s EGD video and Lanza scoring was performed by JLG after the endoscopist had performed pre-treatment and day 7 endoscopies on at least two subjects.

Endpoints

The primary endoscopic endpoint was the day 7 mean total modified gastroduodenal Lanza score (Table 1) for naproxen etemesil vs. naproxen. There was also a prespecified subgroup analysis of the primary endpoint by H. pylori antibody status. The two treatment groups were compared at day 7 for five secondary endoscopic endpoints: total modified gastric Lanza score, percentage of subjects with modified Lanza score ≥5 (≥11 erosions or ulcer) (gastric and gastroduodenal), and percentage of subjects with modified Lanza score = 7 (ulcer; mucosal breaks measuring ≥3 mm in maximum diameter, with depth) (gastric and gastroduodenal). Post hoc analyses included the total number of gastric ulcers and the percentage of subjects with gastric ulcers measuring ≥5 mm in maximum diameter, with depth. Overall safety was assessed by physical examination, laboratory tests, and observed or reported adverse events. Adverse events were coded according to the Medical Dictionary for Regulatory Activities (MedDRA), Version 11.1. Tolerability to study treatments was measured via the Severity of Dyspepsia Assessment (SODA) questionnaire, a validated self-reported dyspepsia outcome measure consisting of an abdominal pain intensity subscore (range 2–47), a non-pain GI symptoms subscore (range 7–35), and a satisfaction subscore (range 2–23).31, 32 Subjects completed the SODA questionnaire on day 1 (pre-treatment), day 8 (post-treatment), and day 15 (end of study follow-up visit).

Statistical analyses

All randomized subjects who received at least one dose of study medication were included in the safety analysis. Subjects completing one post-dosing endoscopy were included in the endoscopy population. Descriptive statistics included means, medians, standard deviation and range. The two-sample t-test assuming unequal variances was used to compare treatment group means for continuous variables and the Cochran-Mantel-Haenszel chi-square test was used to assess treatment group differences for categorical variables. The Benjamini Hochberg correction was used to adjust for multiple comparisons among the secondary gastric and gastroduodenal Lanza score endpoints and to control for false discovery rates. No imputation was performed for missing data. Adverse events were tabulated to examine their frequency, severity, organ systems affected and relationship to study treatment. The SODA questionnaire responses for the safety population were normalized following a validated algorithm. Changes from baseline to day 8 and day 15 in the three separate subscores of SODA were calculated and summarized for each treatment group. Intrasubject changes from baseline were evaluated by treatment group and the magnitudes and directions of the changes were compared between treatment groups. All statistical tests were two-tailed and the significance threshold was set at 5% level. sas for Windows version 9.1.3 was used for all analyses.

The sample size was based on the primary endpoint, and calculated using results from previous similarly designed studies in healthy elderly subjects.27, 28 The reported mean ± standard deviation of the total modified gastroduodenal Lanza score after 1 week of naproxen therapy in 59 healthy subjects aged 65–75 years (40%H. pylori antibody positive) was approximately 3.6 ± 2.728 and in 60 subjects aged 65–76 years (55%H. pylori antibody positive) was approximately 3.0 ± 2.7.27 Using an averaged mean Lanza score of 3.3 ± 2.7 for the naproxen arm, a sample size of 60 subjects per treatment group was calculated to provide 80% power (two-sided; P < 0.05) to detect a 40% lower day 7 total modified gastroduodenal Lanza score for naproxen etemesil (mean difference of −1.32 ± 2.7) compared with naproxen.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Subjects

The trial was conducted from October 2008 through February 2009. A total of 266 subjects were screened, 146 of whom were screen failures (28 had a baseline Lanza score >2, 104 did not meet other inclusion/exclusion criteria, and 14 withdrew consent or were screening back-up subjects determined not to be needed for the study) (Figure 2). The remaining 120 subjects were stratified by H. pylori antibody status and randomized to treatment with either naproxen etemesil (= 61) or naproxen (= 59). Of these, 118 subjects constituted the endoscopy analysis population. The remaining two subjects (both in the naproxen arm) stopped treatment early – one discontinued from the study before day 7 due to moderate abdominal pain and mild diarrhoea and did not have an end-of-study endoscopy; one withdrew consent for the study at the day 7 endoscopy when food was observed and the procedure was aborted.

image

Figure 2.  Subject disposition.

Download figure to PowerPoint

As shown in Figure 2, after completing the day 7 endoscopy, 15 subjects withdrew from treatment and did not participate in the day 8 pharmacokinetic testing. These 15 subjects included 11 subjects with ulcer (naproxen etemesil, = 2; naproxen, = 9) who were required by protocol to stop treatment as well as four additional subjects (two in each arm) who chose to stop treatment early because of endoscopic findings of erosive gastritis or duodenitis. However, these 15 subjects continued in the study for safety follow-up and completed the trial at day 15. The remaining 103 subjects without ulcer (naproxen etemesil, = 57; naproxen, = 46) continued into the pharmacokinetic portion of the trial and completed the trial at day 15.

As shown in Table 2, baseline characteristics and demographics were similar between the two treatment groups, including baseline total modified gastroduodenal Lanza score and the percentage of H. pylori antibody positive subjects. There was a trend towards more females in the naproxen etemesil group (62%) than in the naproxen group (54%).

Table 2.   Baseline characteristics for all randomized subjects
CharacteristicNaproxen etemesil (= 61)Naproxen (= 59)
  1. BMI, body mass index; SODA, severity of dyspepsia assessment.

Age, years
 Mean (s.d.)51.6 (5.3)50.3 (4.1)
 Range45.0–70.645.0–60.0
Age ≥65 years2 (0.3)0
Females, n (%)38 (62.3)32 (54.2)
Race, n (%)
 Caucasian56 (91.8)54 (91.5)
 Black5 (8.2)4 (6.8)
 Other01 (1.7)
Region, n (%)
 Czech Republic34 (55.7)34 (57.6)
 United States27 (44.3)25 (42.4)
Helicobacter pylori antibody positive, n (%)28 (45.9)26 (44.1)
BMI (kg/m2), mean (s.d.)25.2 (2.9)25.7 (2.4)
Gastroduodenal Lanza score, mean (s.d.)0.6 (0.8)0.5 (0.7)
Gastric Lanza score, mean (s.d.)0.5 (0.7)0.4 (0.7)
SODA abdominal pain subscore, mean (s.d.)4.1 (5.3)3.8 (4.7)

Endoscopic endpoints

At baseline, the mean total modified gastroduodenal Lanza score was 0.6 ± 0.8 in the naproxen etemesil group and 0.5 ± 0.7 in the naproxen group. The primary endpoint of mean total modified gastroduodenal Lanza score on day 7 was significantly lower in subjects treated with naproxen etemesil vs. naproxen (2.8 ± 1.7 vs. 3.5 ± 2.0, respectively; = 0.03) (Figure 3). When the primary endpoint was evaluated in a subgroup analysis by H. pylori antibody status, a significant difference remained between treatments for the larger subgroup of patients (= 65) who were H. pylori antibody negative. In these subjects, mean gastroduodenal Lanza score increased from 0.5 ± 0.8 at baseline to 2.5 ± 2.0 on day 7 for naproxen etemesil (= 33), and from 0.4 ± 0.6 at baseline to 3.8 ± 1.9 at day 7 for naproxen (= 32); = 0.01 for the difference between treatments at day 7. There was no significant difference between treatments for the primary endpoint in the smaller H. pylori antibody positive subgroup (= 53), with day 7 scores of 3.0 ± 1.2 for naproxen etemesil (= 28) and 3.2 ± 2.1 for naproxen (= 25); = 0.74.

image

Figure 3.  Box and whisker plot of total modified gastroduodenal Lanza score. Mean total modified gastroduodenal Lanza score on day 7 was significantly less for naproxen etemesil-treated subjects compared with naproxen-treated subjects (2.8 vs. 3.5, respectively; = 0.03).

Download figure to PowerPoint

The secondary endpoints also showed significant differences between treatments favouring lower injury scores for naproxen etemesil (Table 3). The mean modified gastric Lanza score on day 7 was 2.2 ± 1.7 for naproxen etemesil and 3.2 ± 2.2 for naproxen (= 0.005), and the percentage of subjects with a modified gastric Lanza score ≥5 (≥11 erosions or an ulcer) was 6.6% for naproxen etemesil vs. 22.8% for naproxen (= 0.01). Significantly (= 0.02) fewer subjects treated with naproxen etemesil (3.3%, 2/61) developed gastric ulcers (≥3 mm with depth) than subjects treated with naproxen (15.8%, 9/57) (Figure 4). Two subjects in the naproxen group developed at least one duodenal ulcer, as compared with no subject in the naproxen etemesil group. In a post hoc analysis, the total number of gastric ulcers was lower in the naproxen etemesil group (three ulcers in two subjects) than in the naproxen group (29 ulcers in 9 subjects) (Figure 5). Using a more stringent definition of gastric ulcer (mucosal break ≥5 mm in diameter, with depth), results from a post hoc analysis showed the incidence of subjects with gastric ulcer was 0% for naproxen etemesil vs. 12.3% (7/57) for naproxen; = 0.005 (Figure 4). These seven naproxen-treated subjects had a total of 11 ulcers ≥5 mm in length (Figure 5).

Table 3.   Secondary endoscopy endpoints at day 7
Secondary endpointsNaproxen etemesil (= 61)Naproxen (= 57)P-value
  1. * ≥11 erosions or ulcer.

  2. † Two naproxen subjects with gastric ulcer(s) also developed duodenal ulcer(s).

  3. ‡ = N.S. when corrected for multiplicity.

Gastric Lanza score, mean (s.d.)2.2 (1.7)3.2 (2.2)0.005
Subjects with gastric Lanza score ≥5*, n (%)4 (6.6%)13 (22.8%)0.012
Subjects with gastric ulcer, n (%)2 (3.3%)9 (15.8%)0.020
Subjects with gastroduodenal ulcer, n (%)2 (3.3%)9 (15.8%)†0.020‡
Subjects with gastroduodenal Lanza score ≥5*, n (%)8 (13.1%)13 (22.8%)N.S.
image

Figure 4.  Percentage of subjects with gastric ulcer. Significantly fewer subjects in the naproxen etemesil treatment group developed gastric ulcer [≥3 mm (= 0.020) and ≥5 mm (= 0.005)] compared with subjects in the naproxen treatment group.

Download figure to PowerPoint

image

Figure 5.  Total number of gastric ulcers per treatment group. Fewer gastric ulcers (≥3 mm and ≥5 mm) developed in the naproxen etemesil treatment group compared with the naproxen treatment group.

Download figure to PowerPoint

As described elsewhere, plasma pharmacokinetic data demonstrated equivalent exposure to naproxen at steady-state for the naproxen etemesil and naproxen groups.25

Safety and tolerability

All randomized subjects (= 120) received at least one dose of study medication and were included in the safety population. A total of 16 subjects (4/61, 6.6% in the naproxen etemesil group and 12/59, 20.3% in the naproxen group) stopped study treatment early due to treatment emergent adverse events (AEs), including the 11 subjects who developed ulcers (two from the naproxen etemesil group and nine from the naproxen group). The other five subjects stopped treatment due to endoscopic findings of erosive gastritis (one subject in each treatment group), erosive duodenitis (one subject in each treatment group), and moderate abdominal pain and mild diarrhoea (one naproxen subject) (Figure 2).

At least one treatment emergent AE was reported by 63% of naproxen-treated subjects (37/59) and by 53% of naproxen etemesil-treated subjects (32/61). The most commonly reported AEs were GI-related, with 51% of the naproxen group (30/59) and 39% of the naproxen etemesil group (24/61) experiencing a treatment-emergent GI AE (Table 4). The most commonly reported non-endoscopy GI AEs included flatulence, nausea, and abdominal pain or stomach discomfort (Table 4). Most (80%) of treatment emergent AEs were of mild intensity, and only 2% were reported as severe. The severe AEs were increased liver function tests in one naproxen subject and intermittent abdominal discomfort without abnormality on EGD examination in another naproxen subject. There were no reports of death or other serious or significant AEs. There were also no clinically meaningful changes in laboratory parameters.

Table 4.   Most commonly reported non-endoscopy GI adverse events*
Adverse eventNaproxen etemesil (= 61)Naproxen (= 59)
Number of subjects (%)
  1. * Events occurring in ≥5% of subjects.

Flatulence5 (8.2)1 (1.7)
Nausea3 (4.9)3 (5.1)
Stomach discomfort2 (3.3)4 (6.8)
Abdominal pain upper2 (3.3)3 (5.1)
Abdominal pain1 (1.6)3 (5.1)
Vomiting1 (1.6)3 (5.1)

At baseline, there was no difference between the treatment groups in any SODA questionnaire subscore. On day 8, the mean abdominal pain intensity subscore was 6.4 ± 6.6 for naproxen etemesil, and 7.7 ± 8.3 for naproxen. The mean change from baseline to day 8 for naproxen etemesil was 2.2 ± 6.1, and for naproxen 3.9 ± 8.2; these changes from baseline were statistically significant ( 0.008) for each treatment arm, but the difference between the two treatment groups was not statistically significant (= 0.19).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

The pathogenesis of NSAID-associated gastric damage is complex and appears to be influenced by multiple factors. A prevailing theory suggests that systemic COX-mediated inhibition of prostaglandin synthesis leads to a propensity for mucosal injury and delayed mucosal repair.10, 12–14, 18, 33, 34 Topical effects of anti-inflammatory agents, as a physical irritant or through local non-prostaglandin-mediated injury, have also been forwarded as an important mechanism.10–16, 18, 33, 34 This may especially be true for acidic NSAIDs such as naproxen, where direct topical injury is in part predicted by the physicochemical properties of the individual NSAID, resulting in disruption of the mucus-bicarbonate-phospholipid barrier.10, 11 Naproxen etemesil is a lipophilic, non-acidic ester prodrug of naproxen; it derives its anti-inflammatory properties, including nonselective inhibition of COX-1 and COX–2, from the pharmacodynamic actions of the released naproxen. Following oral administration, naproxen etemesil is absorbed intact, followed by hydrolysis to pharmacologically active naproxen. As such, pharmacologically active naproxen does not come into direct contact with the gastrointestinal mucosa. We therefore hypothesized that naproxen etemesil would be associated with less UGI mucosal injury than naproxen.

In this first proof-of-concept evaluation in healthy human volunteers, we compared the gastroduodenal mucosal effects of naproxen etemesil with naproxen, and found that there was less UGI mucosal injury, fewer ulcers and lesser erosive disease in the naproxen etemesil-treated group. The rate of gastric ulcer was 79% lower for subjects in the naproxen etemesil group vs. the naproxen group, with a relative risk = 0.21 (Figure 4).

Two major mechanisms are involved in NSAID-induced gastric mucosal injury – direct local topical injury and systemic prostaglandin inhibition. As such, exposure to the NSAID causes direct topical mucosal damage that is then exacerbated or prevented from healing by the effects of inhibition of mucosal prostaglandin synthesis.10, 12–15 This concept is supported by animal model studies evaluating NSAID-induced injury. COX-1 knockout mice do not spontaneously develop ulcers, yet a single dose of aspirin resulted in severe ulceration, which most likely occurs as a result of local (topical) injury as no COX-1 is present to inhibit.35 The significance of topical injury is further supported in humans in a randomized trial in healthy subjects who received either oral or rectal naproxen 500 mg twice daily.36 Maximal gastric damage was observed in the first 24 h following oral dosing, while no macroscopic gastric damage or significant changes in antral or body mucosal blood flow was observed after rectal administration,36 although these dosing routes have been shown to achieve similar circulating levels of naproxen.37

The present proof-of-concept trial was exploratory, and the results should be interpreted in light of several study limitations. Since the study was conducted in healthy volunteers on a restricted-fat diet, our findings may not be applicable to the chronic arthritis population, who may have specific risk factors for developing NSAID-induced GI mucosal injury. This was partially addressed, however, as our study population had a mean age of 50 years and we stratified the subjects by H. pylori antibody status at baseline. Another potential limitation was the difference in pharmacokinetic profiles between naproxen etemesil, a prodrug from which naproxen is released by hydrolysis, and administered naproxen, which may have introduced bias by the day 7 endoscopy. In the naproxen etemesil group, the observed lower rate of mucosal injury including ulceration occurred despite equivalent steady-state exposure to naproxen as provided by Naprosyn. However, based on a longer half-life for naproxen as derived from naproxen etemesil, steady-state was achieved more slowly with naproxen etemesil, and therefore the cumulative systemic exposure to naproxen after 7 days may have been less for some naproxen etemesil-treated subjects. As such, it is possible that a difference in cumulative systemic drug exposure may, in part, explain the differences observed in this trial. However, steady-state exposure on day 8 was equivalent and the absolute difference in cumulative exposure is unlikely to explain the magnitude of difference observed in ulcer rates (Figure 4).

In addition, the present study was a short-term endoscopic evaluation. Short-term (1-week) endoscopic studies in healthy subjects are a reasonable initial step as proof-of-concept for new therapeutic agents designed to evaluate GI safety.27, 28, 38–40 We found that the percentage of subjects with endoscopically proven gastric ulcer was significantly less for naproxen etemesil compared with naproxen. However, while short-term endoscopic findings provide a preliminary insight into the possible GI safety of a therapeutic agent, well-designed long-term endoscopic trials in subjects requiring NSAID therapy are needed to confirm our findings. While there is less than universal acceptance in the literature as to whether endoscopic ulcers, regardless of trial duration, are a meaningful surrogate for clinically significant GI outcomes such as bleeding and perforation, this study provides evidence of the reduced rate of mucosal injury for naproxen etemesil as compared with naproxen.41–43

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

The pathogenesis of NSAID-induced GI mucosal injury is multifactorial. In addition to the prostaglandin-independent (local) and -dependent (systemic) mechanisms, the physicochemical and pharmacokinetic properties of the individual NSAID, the dose of the NSAID, and individual patient risk factors play a role. The results of our pilot study in healthy volunteers with a mean age of 50 years are encouraging in that we found after 7 days of exposure, naproxen etemesil (an inactive, non-acidic prodrug of naproxen) was associated with significantly less UGI mucosal injury compared with naproxen. These findings require validation in long-term endoscopic trials in patients requiring chronic NSAID therapy.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Declaration of personal interests: Dr Jay L. Goldstein has received research and/or educational funding, consulting fees, contract payments and speaker’s honoraria from AstraZeneca, Given, Horizon, Novartis, Pfizer, POZEN, Takeda and TAP. He has received consulting fees from Amgen, Astellas, GlaxoSmithKline, Merck, Novartis, PLX, Procter & Gamble and Wyeth. Grants have been awarded from Amgen, Novartis and GlaxoSmithKline and contract payments from Amgen and GlaxoSmithKline. For Logical Therapeutics, he was a study investigator and GI trainer for this clinical trial. In addition, Dr Goldstein has received consulting fees from Logical Therapeutics. Drs Anna Jungwirthová, Joseph David, Enrique Spindel, Luděk Bouchner, František Pešek, Shawn Searle, JiříŠkopek, Jiří Grim, and Ivan Ulč have no conflicts to declare. Dr K. Lea Sewell is an employee of Logical Therapeutics, Inc. Declaration of funding interests: The study was funded in full by Logical Therapeutics, Inc. Statistical support was provided by Pete Shabe, PhD and Don Young, PhD of Advance Research Associates, Inc. and funded by Logical Therapeutics, Inc. Writing support was provided by Lorraine R. Baer, PharmD of Baer PharMed Consulting, Ltd, and funded by Logical Therapeutics, Inc.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  • 1
    Elnachef N, Scheiman JM, Fendrick AM, Howden CW, Chey WD. Changing perceptions and practices regarding aspirin, nonsteroidal anti-inflammatory drugs, and cyclooxygenase-2 selective nonsteroidal anti-inflammatory drugs among US primary care providers. Aliment Pharmacol Ther 2008; 28: 124958.
  • 2
    Laine L. Gastrointestinal effects of NSAIDs and coxibs. J Pain Symptom Manage 2003; 25(2 Suppl): S3240.
  • 3
    Cryer B. NSAID-associated deaths: the rise and fall of NSAID-associated GI mortality. Am J Gastroenterol 2005; 100: 16945.
    Direct Link:
  • 4
    Lanas A, Perez-Aisa MA, Feu F, et al. A nationwide study of mortality associated with hospital admission due to severe gastrointestinal events and those associated with nonsteroidal antiinflammatory drug use. Am J Gastroenterol 2005; 100: 168593.
    Direct Link:
  • 5
    Straube S, Tramèr MR, Moore RA, Derry S, McQuay HJ. Mortality with upper gastrointestinal bleeding and perforation: effects of time and NSAID use. BMC Gastroenterol 2009; 9: 41.
  • 6
    Jones R, Rubin G, Berenbaum F, Scheiman J. Gastrointestinal and cardiovascular risks of nonsteroidal anti-inflammatory drugs. Am J Med 2008; 121: 46474.
  • 7
    Lanas A. A review of the gastrointestinal safety data – a gastroenterologist’s perspective. Rheumatology (Oxford) 2010; 49(Suppl 2): ii310.
  • 8
    García Rodríguez LA, Tolosa LB. Risk of upper gastrointestinal complications among users of traditional NSAIDs and COXIBs in the general population. Gastroenterology 2007; 132: 498506.
  • 9
    Hernández-Díaz S, Rodríguez LA. Association between nonsteroidal anti-inflammatory drugs and upper gastrointestinal tract bleeding/perforation: an overview of epidemiologic studies published in the 1990s. Arch Intern Med 2000; 160: 20939.
  • 10
    Musumba C, Pritchard DM, Pirmohamed M. Review article: cellular and molecular mechanisms of NSAID-induced peptic ulcers. Aliment Pharmacol Ther 2009; 30: 51731.
  • 11
    Bjarnason I, Scarpignato C, Takeuchi K, Rainsford KD. Determinants of the short-term gastric damage caused by NSAIDs in man. Aliment Pharmacol Ther 2007; 26: 95106.
  • 12
    Laine L, Takeuchi K, Tarnawski A. Gastric mucosal defense and cytoprotection: bench to bedside. Gastroenterology 2008; 135: 4160.
  • 13
    Rich M, Scheiman JM. Nonsteroidal anti-inflammatory drug gastropathy at the new millennium: mechanisms and prevention. Semin Arthritis Rheum 2000; 30: 16779.
  • 14
    Scheiman JM. NSAIDs, gastrointestinal injury, and cytoprotection. Gastroenterol Clin North Am 1996; 25: 27998.
  • 15
    Sostres C, Gargallo CJ, Arroyo MT, Lanas A. Adverse effects of non-steroidal anti-inflammatory drugs (NSAIDs, aspirin and coxibs) on upper gastrointestinal tract. Best Pract Res Clin Gastroenterol 2010; 24: 12132.
  • 16
    Ham M, Kaunitz JD. Gastroduodenal mucosal defense. Curr Opin Gastroenterol 2007; 23: 607.
  • 17
    Cryer B, Feldman M. Cyclooxygenase-1 and cyclooxygenase-2 selectivity of widely used nonsteroidal anti-inflammatory drugs. Am J Med 1998; 104: 41321.
  • 18
    Petruzzelli M, Vacca M, Moschetta A, et al. Intestinal mucosal damage caused by non-steroidal anti-inflammatory drugs: role of bile salts. Clin Biochem 2007; 40: 50310.
  • 19
    Rainsford KD. Profile and mechanisms of gastrointestinal and other side effects of nonsteroidal anti-inflammatory drugs (NSAIDs). Am J Med 1999; 107(6A): 27S35S.
  • 20
    Giraud MN, Motta C, Romero JJ, Bommelaer G, Lichtenberger LM. Interaction of indomethacin and naproxen with gastric surface-active phospholipids: a possible mechanism for the gastric toxicity of nonsteroidal anti-inflammatory drugs (NSAIDs). Biochem Pharmacol 1999; 57: 247.
  • 21
    Lichtenberger LM, Zhou Y, Dial EJ, Raphael RM. NSAID injury to the gastrointestinal tract: evidence that NSAIDs interact with phospholipids to weaken the hydrophobic surface barrier and induce the formation of unstable pores in membranes. J Pharm Pharmacol 2006; 58: 14218.
  • 22
    Goldstein JL. Challenges in managing NSAID-associated gastrointestinal tract injury. Digestion 2004; 69(Suppl 1): 2533.
  • 23
    Goldstein JL, Hochberg MC, Fort JG, Zhang Y, Hwang C, Sostek M. Clinical trial: incidence of NSAID-associated endoscopic gastric ulcers in patients treated with PN400 (naproxen plus esomeprazole magnesium) vs. enteric-coated naproxen alone. Aliment Pharmacol Ther 2010; 32: 40113.
  • 24
    Laine L, Schiff M, Genovese M. Does high-dose famotidine reduce gastric and duodenal ulcers in NSAID users? Two double-blind six-month trials of single-tablet combination ibuprofen-famotidine vs. ibuprofen alone (REDUCE-1 and 2) [abstract]. Gastroenterology 2009; 136(5 Suppl 1): A69.
  • 25
    Goldstein JL, Kramer WG, Searle S, et al. LT-NS001, a naproxen pro-drug with minimal COX inhibition while intact, results in equal plasma exposure to naproxen in volunteers aged 45–70 compared with Naprosyn®. Gut 2009; 58(Suppl II): A270.
  • 26
    Lanza FL, Royer GL Jr, Nelson RS. Endoscopic evaluation of the effects of aspirin, buffered aspirin, and enteric-coated aspirin on gastric and duodenal mucosa. N Engl J Med 1980; 303: 1368.
  • 27
    Goldstein JL, Kivitz AJ, Verburg KM, Recker DP, Palmer RC, Kent JD. A comparison of the upper gastrointestinal mucosal effects of valdecoxib, naproxen and placebo in healthy elderly subjects. Aliment Pharmacol Ther 2003; 18: 12532.
  • 28
    Goldstein JL, Aisenberg J, Lanza F, et al. A multicenter, randomized, double-blind, active-comparator, placebo-controlled, parallel-group comparison of the incidence of endoscopic gastric and duodenal ulcer rates with valdecoxib or naproxen in healthy subjects aged 65 to 75 years. Clin Ther 2006; 28: 34051.
  • 29
    Yeomans ND, Næsdal J. Systematic review: ulcer definition in NSAID ulcer prevention trials. Aliment Pharmacol Ther 2008; 27: 46572.
  • 30
    Taha AS, Hudson N, Hawkey CJ, et al. Famotidine for the prevention of gastric and duodenal ulcers caused by NSAIDs. N Engl J Med 1996; 34: 14359.
  • 31
    Rabeneck L, Cook KF, Wristers K, Souchek J, Menke T, Wray NP. SODA (severity of dyspepsia assessment): a new effective outcome measure for dyspepsia-related health. J Clin Epidemiol 2001; 54: 75565.
  • 32
    Rabeneck L, Wristers K, Goldstein JL, Eisen G, Dedhiya SD, Burke TA. Reliability, validity, and responsiveness of severity of dyspepsia assessment (SODA) in a randomized clinical trial of a COX-2-specific inhibitor and traditional NSAID therapy. Am J Gastroenterol 2002; 97: 329.
    Direct Link:
  • 33
    Lichtenberger LM. Where is the evidence that cyclooxygenase inhibition is the primary cause of nonsteroidal anti-inflammatory drug (NSAID)-induced gastrointestinal injury? Topical injury revisited. Biochem Pharmacol 2001; 61: 6317.
  • 34
    Tomisato W, Tsutsumi S, Hoshino T, et al. Role of direct cytotoxic effects of NSAIDs in the induction of gastric lesions. Biochem Pharmacol 2004; 67: 57585.
  • 35
    Darling RL, Romero JJ, Dial EJ, Akunda JK, Langenbach R, Lichtenberger LM. The effects of aspirin on gastric mucosal integrity, surface hydrophobicity, and prostaglandin metabolism in cyclooxygenase knockout mice. Gastroenterology 2004; 127: 94104.
  • 36
    Lipscomb GR, Rees WD. Gastric mucosal injury and adaptation to oral and rectal administration of naproxen. Aliment Pharmacol Ther 1996; 10: 1338.
  • 37
    Desager JP, Vanderbist M, Harvengt C. Naproxen plasma levels in volunteers after single-dose administration by oral and rectal routes. J Clin Pharmacol 1976; 16: 18993.
  • 38
    Lanza FL, Rack MF, Simon TJ, et al. Specific inhibition of cyclooxygenase-2 with MK-0966 is associated with less gastroduodenal damage than either aspirin or ibuprofen. Aliment Pharmacol Ther 1999; 13: 7617.
  • 39
    Rordorf C, Kellett N, Mair S, et al. Gastroduodenal tolerability of lumiracoxib vs placebo and naproxen: a pilot endoscopic study in healthy male subjects. Aliment Pharmacol Ther 2003; 18: 53341.
  • 40
    Simon LS, Lanza FL, Lipsky PE, et al. Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor: efficacy and safety in two placebo-controlled trials in osteoarthritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum 1998; 41: 1591602.
  • 41
    Cryer B. Reducing the gastrointestinal risks of low-dose aspirin. Gastroenterology 2010; 38: 303.
  • 42
    Moore A, Bjarnason I, Cryer B, et al. Evidence for endoscopic ulcers as meaningful surrogate endpoint for clinically significant upper gastrointestinal harm. Clin Gastroenterol Hepatol 2009; 7: 115663.
  • 43
    Graham DY. Endoscopic ulcers are neither meaningful nor validated as a surrogate for clinically significant upper gastrointestinal harm. Clin Gastroenterol Hepatol 2009; 7: 114750.