The new proton pump inhibitor esomeprazole is effective as a maintenance therapy in GERD patients with healed erosive oesophagitis: a 6-month, randomized, double-blind, placebo-controlled study of efficacy and safety
N. B. Vakil,
University of Wisconsin Medical School, Milwaukee, WI, USA
Dr N. B. Vakil, University of Wisconsin Medical School, Sinai Samaritan Medical Center, 945 N 12th Street, Room 4040, Milwaukee, WI 53233, USA. E-mail: firstname.lastname@example.org
Esomeprazole, the S-isomer of omeprazole, is the first proton pump inhibitor to be developed as an optical isomer. In patients with erosive oesophagitis, esomeprazole has produced significantly greater healing rates and improved symptom resolution vs. omeprazole.
This study assesses the efficacy of esomeprazole for preventing relapse in patients with healed oesophagitis.
In this 6-month US multicentre randomized double-blind placebo-controlled trial, 375 Helicobacter pylori-negative patients with endoscopically healed oesophagitis received esomeprazole 40 mg, 20 mg, 10 mg, or placebo once daily. The primary efficacy end-point was maintenance of healing at 6 months. Secondary end-points assessed changes in symptoms, and long-term safety and tolerability.
Significantly (P < 0.001) more patients remained healed with esomeprazole 40 mg (87.9%), 20 mg (78.7%), or 10 mg (54.2%), than with placebo (29.1%). Relapse, when it occurred, was later with esomeprazole. Sustained resolution of heartburn was observed in the 40 mg and 20 mg groups; there was a high correlation between absence of heartburn and maintenance of healing. Adverse effects were mild, infrequent and not significantly different between groups.
Esomeprazole is effective and well-tolerated in the maintenance of healing of erosive oesophagitis. Esomeprazole 40 mg and 20 mg offer significant clinical benefit to patients.
Erosive oesophagitis is a chronic, recurring disease. Approximately 75% of patients previously diagnosed with oesophagitis still have significant morbidity related to reflux disease more than 10 years after diagnosis.1 Following successful initial treatment of oesophagitis, at least 50% of patients relapse within 6–12 months after drug therapy is withdrawn; some studies report relapse rates of up to 80–90%.2–7 Thus, patients generally require long-term antisecretory therapy following initial healing. The rate of relapse is markedly reduced after effective maintenance treatment with proton pump inhibitors; relapse rates drop to approximately 10% to 30% with once daily omeprazole 20 mg or lansoprazole 30 mg in the first year.7 Proton pump inhibitors are currently the agents of choice both to treat erosive oesophagitis and to maintain healing.8
Esomeprazole, the S-isomer of omeprazole, has been developed to further improve pharmacotherapy of gastro-oesophageal reflux disease (GERD). It is the first proton pump inhibitor to be developed as an optical isomer. Pharmacokinetic and pharmacodynamic studies indicate advantages of esomeprazole over omeprazole: because of slower clearance, a 40-mg or 20-mg dose of esomeprazole exhibits a significantly higher area under the plasma concentration–time curve (AUC) than a 20-mg dose of omeprazole.9 This results in a greater and more consistent acid suppression.9,10 These pharmacologic advantages translate into clinical benefits in patients with acid-related diseases. Esomeprazole 40 mg o.d. produced consistently high healing rates (94.1% and 93.7% at week 8) and sustained resolution of heartburn (64.7% and 68.3% at week 4, investigator assessment) in two 8-week randomized controlled trials evaluating the initial treatment of active erosive oesophagitis. In these trials, healing rates and symptom resolution (seven consecutive heartburn-free days) were significantly higher with esomeprazole than those achieved with omeprazole 20 mg (the approved and standard dose for treating erosive oesophagitis).11,12
The aim of the current trial was to determine the efficacy of esomeprazole in maintenance of healing of erosive oesophagitis. The primary end-point was determination of the proportion of patients who had maintenance of healing of erosive oesophagitis after 6 months of maintenance therapy with esomeprazole 40 mg, 20 mg and 10 mg. Secondary end-points evaluated the efficacy of esomeprazole in the control of GERD symptoms, and the long-term safety and tolerability of esomeprazole.
Study design and patient selection
This multicentre, double-blind, randomized, placebo-controlled, parallel-group study was conducted at 51 centres in the US. Enrolled patients were Helicobacter pylori-negative (determined by histology) and had endoscopically confirmed healing of erosive oesophagitis (Los Angeles [LA] classification ‘not present’)13 after an acute healing trial that immediately preceded the maintenance trial. The healing study was an 8-week multicentre double-blind randomized parallel group trial that compared efficacy and tolerability of esomeprazole 40 mg, esomeprazole 20 mg, and omeprazole 20 mg administered once daily in 1960 patients with endoscopically confirmed erosive oesophagitis.11 Based on power calculations, a smaller number of patients was required for the maintenance study; therefore only some of the centres that participated in the acute healing study participated in this maintenance study.
Gastric biopsy results and histologically determined H. pylori status measured at entry to the healing study were used as the baseline values for this maintenance study. Results from the final visit of the healing study served as baseline values for medical history, physical examination, and blood and urine samples. Endoscopy results from the final visit of the healing study also served as baseline efficacy assessments for this study. Females of child-bearing potential had urine dipstick and serum pregnancy tests repeated at baseline of the maintenance study.
Patients were randomized to one of four treatment groups: esomeprazole 40 mg (n=92); esomeprazole 20 mg (n=98); esomeprazole 10 mg (n=91); and placebo (n=94). All study drugs were identical in appearance to maintain blinding, and were taken orally once daily in the morning. Treatment was allocated using individually sealed and blinded randomization envelopes that were collected and checked by the monitor at the end of the study to ensure the integrity of blinding. Rescue medication (aluminium/magnesium hydroxide, Gelusil tablets; Warner- Lambert Consumer Healthcare, Morris Plains, NJ) was provided for acute relief of symptoms of GERD, up to a maximum of six tablets per day. Treatment compliance and use of rescue medication were assessed by capsule/tablet count at each visit.
Return visits were scheduled at 1, 3, and 6 months, at which times GERD symptoms were assessed and an esophagogastroduodenoscopy was performed. If oesophagitis (i.e. LA Classification Grade of A, B, C, or D) was present at any visit, the patient was considered to have relapsed and was discontinued from the study. At the final visit, gastric biopsies were repeated for histological evaluation.
The study was conducted in accordance with the Declaration of Helsinki and in compliance with Good Clinical Practices regulations and guidance issued by the Food and Drug Administration. Institutional review board approval was obtained at each study site. All patients provided written informed consent.
Males and non-pregnant, non-lactating females, between the ages of 18 and 75 years inclusive, who had confirmed healing of erosive oesophagitis, no record of any serious adverse event related to study medication in the healing study, and who were negative for H pylori by histology were eligible for entry into this maintenance study. Females were required to be post-menopausal, surgically sterilized, or using a medically acceptable form of birth control throughout the study.
Patients with endoscopic Barrett’s oesophagus (> 3 cm) or significant dysplastic changes in the oesophagus were excluded, as were patients with current or historical evidence (within 3 months) of significant gastrointestinal disease or any other significant medical condition or disease. Other exclusion criteria included: use of a proton pump inhibitor (other than as part of the protocol of the healing study) within 28 days prior to the baseline visit; and use of an H2-receptor antagonist daily during the 2 weeks prior to the baseline esophagogastroduodenoscopy (occasional use less than daily during these 2 weeks was permitted). Additionally, patients taking non-steroidal anti-inflammatory drugs, quinidine, diazepam, diphenylhydantoin, mephenytoin, warfarin, anticholinergics, prostaglandin analogs, antineoplastic agents, salicylates (unless ≤ 165 mg daily), oral or intravenous steroids, pro-motility drugs, sucralfate, and those with a known hypersensitivity to esomeprazole or aluminium/magnesium hydroxide were excluded.
The primary efficacy end-point was the proportion of patients in endoscopic remission from baseline through to month 6. Maintenance of healing was defined as the absence of erosions (LA Classification Grade ‘not present’) based on evaluation by esophagogastroduodenoscopy. The LA Classification was used because of the high level of inter-observer reliability it allows among experienced endoscopists.13,14
Secondary efficacy end-points included a sub-group analysis of 6-month maintenance of healing rates based on patient demographics: gender; age (< 65 years, ≥ 65 years); race; LA Grade at entry into healing study; treatment received during the healing study; duration of treatment in healing study (4 or 8 weeks); and investigational site. The mean time to recurrence of oesophagitis was also calculated. The investigators evaluated the GERD symptom heartburn at baseline (the final visit of the healing trial) and at months 1, 3, and 6. Heartburn was defined as a burning feeling, rising from the stomach or lower part of the chest towards the neck. Investigators recorded the severity of the most intense episode of heartburn within the 7 days immediately preceding the visit using the following rating scheme: none (no symptoms); mild (awareness of symptom, but easily tolerated); moderate (discomforting symptom sufficient to cause interference with normal activities including sleep); and severe (incapacitating symptoms, with inability to perform normal activities including sleep). The percentage of patients with sustained resolution of heartburn (defined as an investigator severity rating of ‘none’ during the previous 7 days) was then determined. The relationship between maintenance of oesophageal healing and the absence of heartburn was also determined.
At each return visit, clinical laboratory evaluations were performed, and any adverse events were reported. Adverse events were classified as to severity and relationship to study drug. The percentage of patients who reported any adverse event, serious adverse event, and treatment-related adverse event, and who discontinued treatment because of an adverse event was determined.
Clinical laboratory evaluations included serum chemistry and haematology. Antral and corporal biopsy samples were histologically evaluated for three characteristics of gastritis (chronic inflammation, atrophy, and intestinal metaplasia) using the updated Sydney System.15 Biopsy samples from the greater curvature of the corpus were evaluated for enterochromaffin-like cell hyperplasia using the eight-point classification scale developed by Solcia et al.16,17
SmithKline Beecham Clinical Laboratories, Inc. (Van Nuys, CA) conducted clinical laboratory testing. Histopathological slides for all biopsy specimens were prepared and evaluated at a central laboratory (Baylor College of Medicine, Houston, TX). All enterochromaffin-like cell hyperplasia evaluations were carried out in a central laboratory (the University of Pavia, Pavia, Italy).
All summary tables and descriptive and inferential statistics were generated using the SAS System, Version 6.12. Efficacy endpoints were analysed for the intention-to-treat population, defined as all patients who were randomized to treatment. The safety analysis included any patient who received at least one documented dose of study medication.
At month 6, cumulative life-table rates were used to estimate the maintenance of healing, and each esomeprazole treatment group was compared with placebo using a log-rank test. For these comparisons, the experiment-wise error rate was preserved at 0.05 using the Hochberg procedure to adjust for multiple comparisons. The presence of a dose–response relationship was evaluated informally using descriptive statistics. Crude rates for the maintenance of healing were calculated by month for patient sub-groups (based on gender, age group, race, initial severity of erosive oesophagitis, treatment used in the healing study, time to healing, and investigational site), but no inferential statistical comparisons were made.
The proportion of patients with heartburn present at month 1 for each esomeprazole treatment group was compared with that for placebo using Cochran–Mantel–Haenszel statistics, adjusting for the absence or presence of heartburn at baseline. Heartburn at month 3 and month 6 was summarized using descriptive statistics only. Concurrence between the maintenance of healing and the absence of GERD symptoms was summarized by month.
Summaries of adverse events, and incidence rates of adverse events by body system and preferred term, were tabulated over the entire study period (through to week 26), and for events occurring through to week 13 (month 3) and through to week 4 (month 1). No inferential statistics were performed on adverse events. Laboratory test results were summarized using descriptive statistics for the change from baseline in each test by visit. Vital signs were examined at each visit using descriptive statistics.
Summaries of gastritis ratings were tabulated for baseline (from the healing study) and final biopsy data. The proportion of patients with an increase in enterochromaffin-like cell rating in each of the esomeprazole treatment groups was compared with the proportion in the placebo group using logistic regression models, with time in the study (≤ 3 months, or > 3 months) included as a covariate.
Patient characteristics and disposition
A total of 375 patients were randomized to one of four treatment groups at 51 study sites in the US. The distribution of patients across treatment groups was generally similar with respect to baseline characteristics, and the majority of patients in all four groups had no heartburn at baseline (Table 1). The distribution of patients across treatment groups was also similar according to LA Classification Grade at entry to the healing study, treatment received, and duration of treatment in the healing study.
Table 1. Baseline characteristics
A total of 191 patients (50.9%) completed the 6-month study. Completion rates increased in direct proportion to the dose of esomeprazole, from 21.3% (n=20) in the placebo group to 72.8% (n=67) in the esomeprazole 40 mg group. Loss to follow-up, sponsor/investigator decision, and withdrawal of consent, caused most of the discontinuations from the esomeprazole 40 mg group (16 out of 25 discontinuations, 64%) and 20 mg group (20 of 36 discontinuations, 56%). In contrast, these were reasons for withdrawal in the minority of the esomeprazole 10 mg group (17 out of 49 discontinuations, 35%) and placebo group (12 out of 74 discontinuations, 16%). Lack of therapeutic response was the most frequent reason for discontinuation from the esomeprazole 10 mg group (30 out of 49 discontinuations, 61%) and placebo group (60 out of 74 discontinuations, 81%), but was less common in the esomeprazole 40 mg group (four out of 25 discontinuations, 16%) and esomeprazole 20 mg group (11 out of 36 discontinuations, 31%). Overall, few patients (n=15) discontinued treatment because of adverse events.
Maintenance of healing of erosive oesophagitis
The rate of maintenance of healing with each dose of esomeprazole was significantly greater (life-table estimates, all P-values < 0.001) than with placebo at each visit (Figure 1). Life table estimates indicated that cumulative healing was maintained at month 6 in 87.9% of patients treated with esomeprazole 40 mg (95% CI: 80.4–95.4%), 78.7% on esomeprazole 20 mg (95% CI: 69.5–87.8%), and 54.2% on esomeprazole 10 mg (95% CI: 42.9–65.5%), compared with 29.1% of placebo patients (95% CI: 17.6–40.6%; P < 0.001; log rank test). While formal statistical comparisons between the esomeprazole treatment groups were not conducted, the maintenance of healing rates revealed a dose–response relationship. Specifically, rates of maintenance of healing at month 6 increased as the dose of esomeprazole increased. In those patients who did experience a recurrence of erosive oesophagitis, the mean time to recurrence was shorter in patients receiving placebo (46 days) than in patients receiving esomeprazole (80, 101, and 130 days for the 10 mg, 20 mg and 40 mg groups, respectively). No patient had a grade D recurrence and few patients recurred at a grade that was higher than their initial grade.
Esomeprazole was effective regardless of the pre-treatment severity of oesophagitis and patient sub-group. Sub-group data indicate that no variable, including gender, age (≥ 65 years vs. < 65 years), healing study treatment, time to healing (4 weeks vs. 8 weeks), and baseline severity of oesophagitis, appeared to predict maintenance of healing by esomeprazole. In contrast, recurrence was more likely in the placebo group in those patients with higher initial grades of oesophagitis.
Heartburn was the most common symptom of GERD in patients enrolled in the healing study, although most patients were symptom-free at the time of entry into this study. At month 1, the proportion of patients who were heartburn-free during the week prior to the visit was highest in the esomeprazole 40 mg group (71.3%, 62 out of 87), lower in the esomeprazole 20 mg (63.7%, 58 out of 91) and 10 mg (50.6%, 42 out of 83) groups, and lowest in the placebo group (15.5%, 13 out of 84). Each pair-wise comparison of esomeprazole with placebo was statistically significant (all P-values < 0.001). The frequencies of GERD symptoms evaluated by the investigator as none (no symptoms) or only mild were 95.4% (83 out of 87), 87.9% (80 out of 91), 85.5% (71 out of 83), and 33.3% (28 out of 84), respectively, in the four groups (esomeprazole 40 mg, 20 mg, 10 mg and placebo). Again, each pair-wise comparison of esomeprazole with placebo was statistically significant (all P-values < 0.001).
Pair-wise comparisons between esomeprazole and placebo for rates of symptom control at month 3 and month 6 must be interpreted cautiously because only patients who maintained healing at the previous visit continued to the subsequent visits. At month 3, heartburn was absent in 68.0% (51 out of 75), 62.0% (49 out of 79), and 53.4% (31 out of 58) of patients remaining in the esomeprazole 40 mg, 20 mg and 10 mg groups, respectively. This was compared with 45.8% (11 out of 24) of patients remaining in the placebo group. At month 6, heartburn was absent in 77.6% (52 out of 67), 77.8% (49 out of 63), and 66.7% (30 out of 45) of the esomeprazole 40 mg, 20 mg and 10 mg groups, respectively, compared with 52.4% (11 out of 21) of the placebo group. A similar trend was seen for patients with none or only mild heartburn.
Maintenance of healing was highly correlated with absence of heartburn. Among patients who reported no heartburn, healing was maintained in more than 98% of the esomeprazole 40 mg and 20 mg groups, 95% of the esomeprazole 10 mg group, and 92% of the placebo group. However, a recurrence of symptoms was not predictive of a relapse of erosive oesophagitis in patients treated with esomeprazole. Only 4.2% of the esomeprazole 40 mg group, 15.2% of the 20 mg group, and 38.5% of the 10 mg group who had heartburn had a recurrence of erosive oesophagitis. In contrast, 69.1% of the placebo group who reported symptoms had a recurrence of erosive oesophagitis. Few patients with recurrence of erosive oesophagitis were heartburn-free in any treatment group (one patient on esomeprazole 40 mg, 1.6%; no patients on esomeprazole 20 mg; two patients on esomeprazole 10 mg, 4.9%; and one patient on placebo, 8.3%).
Greater than 78% of patients in each group had compliance rates greater than 80% over the 6-month treatment period. Mean Gelusil use was 2–3 times higher in the placebo group (0.9–1.8 tablets/day) than in the esomeprazole study groups (0.4–0.9 tablets/day). There did not appear to be any relationship between Gelusil and esomeprazole dose, nor to time in the study.
Esomeprazole recipients remained in the study notably longer than placebo recipients (mean range 116–146 days vs. 61.5 days). Therefore a direct comparison of the incidence of adverse events is difficult. Of the esomeprazole groups, 60.9% of the 40 mg group, 57.1% of the 20 mg group and 57.1% of the 10 mg group experienced at least one adverse event throughout the entire study period in contrast with 45.7% of the placebo group. Most adverse events were mild to moderate, and no serious adverse event was considered by the investigator to be related to treatment. There were no clinically meaningful changes in vital signs or physical examinations. Discontinuation from treatment attributed to adverse events was six (6.5%), five (5.5%) and two (2.2%) patients in the 40 mg, 20 mg, and 10 mg groups, respectively, compared with placebo two placebo recipients (2.2%).
At month 1, the overall proportions of patients reporting at least one adverse event and the percentages of most frequently occurring adverse events were similar in the esomeprazole groups and the placebo group (Table 2). A comparison of adverse event rates between groups at month 6 was not performed, because only 21.7% of placebo patients remained in the trial vs. 72.8%, 63.3%, and 46.2% of esomeprazole 40 mg, 20 mg, and 10 mg recipients, respectively. The most frequently reported adverse events over the entire 6-month study period were headache (7.8%), respiratory infection (6.7%), and diarrhoea (6.4%).
Table 2. Adverse events at month 1
A number of laboratory parameters were studied to explore individual and mean changes from baseline. As expected with prolonged acid suppression, mean gastrin levels increased with higher doses of esomeprazole, with most changes occurring within the first month. Among the placebo patients, all of whom had received therapy with esomeprazole or omeprazole in the healing trial, mean serum gastrin levels returned to the baseline levels recorded prior to entry in the acute healing trial. The increase in mean gastrin level from baseline of the current trial to final visit was 50.4 ng/L (s.d. 74.7 ng/L) in patients receiving esomeprazole 40 mg, and 21.3 ng/L (s.d. 59.8 ng/L) with esomeprazole 20 mg. Mean gastrin levels decreased by 0.71 ng/L (s.d. 65.37 ng/L) in the esomeprazole 10 mg group and 26.21 ng/L (s.d. 56.33 ng/L) in the placebo group. Otherwise, laboratory changes in individuals, whether increases or decreases, tended to be small and the majority of values remained well within normal ranges. Individual changes were not found to be clinically meaningful. Any shifts in population mean were small and did not suggest safety concerns. There was no pattern that indicated a trend.
Gastric biopsy evaluations revealed very few abnormal ratings for the three gastritis characteristics (chronic inflammation, atrophy, or intestinal metaplasia) at either antral or corporal locations and across all biopsy sites. Improvement and worsening of these characteristics were distributed evenly across the four treatment groups and more patients had improvement than worsening in their gastritis ratings. No new occurrences of atrophic gastritis were observed at the final biopsy.
Enterochromaffin-like cell ratings showed that one patient in the esomeprazole 20 mg group had a normal baseline rating and micro-nodular hyperplasia at the final biopsy; all other abnormal, post-baseline enterochromaffin-like cell ratings were either linear hyperplasia (in four patients in the esomeprazole 20 mg group) or simple hyperplasia (in nine, seven, and three patients in the esomeprazole 40 mg, 20 mg and 10 mg groups, respectively, and in one patient in the placebo group). None of these changes was thought to indicate a safety concern, and there was no evidence of adenomatous changes or malignancy.
GERD is a recurring disease and the majority of patients with erosive and non-erosive reflux disease have a recurrence after discontinuation of initial therapy. Indeed, in this study, most placebo patients relapsed within 1–3 months of treatment. Maintenance therapy with a proton pump inhibitor is significantly more effective than standard dose H2-RA therapy for maintaining the healing effect.8 In this study, esomeprazole 40 mg and 20 mg were highly effective for maintaining healing of erosive oesophagitis. Although esomeprazole 10 mg also demonstrated significantly greater efficacy than placebo, this dose was not considered to be clinically viable for this patient population.
A dose response was observed in the study, with esomeprazole 40 mg demonstrating greater efficacy than esomeprazole 20 mg. Both doses of esomeprazole (40 mg and 20 mg) significantly reduced the relapse rate of erosive oesophagitis compared with placebo, such that the absolute risk reduction for each dose vs. placebo was 58.8% and 49.6%, respectively. The absolute risk reduction in this study indicates that the number of patients that would need to be treated (number needed to treat; NNT) with esomeprazole 40 mg and 20 mg for 6 months to prevent one episode of recurrent oesophagitis is 1.7 and 2.0, respectively.
In the current trial, the majority of patients treated with esomeprazole 40 mg or 20 mg remained heartburn free after 1 month of therapy, while heartburn was present in most placebo-treated patients. There was a very high concurrence between the absence of heartburn and maintenance of healing during esomeprazole maintenance therapy. Therefore the control of symptoms can be used as a reliable indicator of maintained healing of oesophagitis. These results support the conclusions of a recent consensus statement by the Genval Group, which concluded that routine endoscopic monitoring is inappropriate and that symptom control provides a reliable indicator of healing.8
Treatment withdrawals from the placebo group because of lack of efficacy prevent a meaningful comparison of adverse events between the esomeprazole groups and the placebo group beyond month 1. At 1 month, adverse events were similar across all treatment groups, including the placebo group, and adverse events reported with esomeprazole during the 6-month period were usually mild and moderate and were those expected of a proton pump inhibitor. The most common adverse events over the study period were headache, respiratory infection, and diarrhoea. These findings are consistent with those of a 12-month open label safety trial designed to expose a sufficient number of patients to esomeprazole (at a dose of 40 mg o.d.) to be able to characterize the general safety profile of the drug in chronic dosing.18
Histology findings from the current trial have been combined with those from a similar randomized double-blind trial to form a pooled analysis of safety data from 519 patients with healed erosive oesophagitis who received esomeprazole. Results from these studies and a 12-month safety study, which evaluated 807 healed erosive oesophagitis patients treated with esomeprazole 40 mg o.d. for up to 12 months, have been reported elsewhere and are consistent with the findings of the current trials.19
In conclusion, the majority of patients with erosive oesophagitis will require long-term care. Long-term therapy with esomeprazole 40 mg or 20 mg maintains healing of erosive oesophagitis in most patients irrespective of baseline severity of oesophagitis. The control of heartburn is a reliable indicator of maintenance of healing.
The Esomeprazole Study Investigators: K. H. Ashby, Pacific Coast Clinical Coordinators; R. A. Baerg, Tacoma Digestive Disease Center; I. Bassan; M. N. Baz, Asthma and Allergy Center; R. W. Bennetts, North-west Gastroenterology Clinic; D. Biery; VA Botoman, Cleveland Clinic (FL); L. B. Cohen; D. W. Collins, Western States Clinical Research; C. Dasher, Norwood Clinic Research Center; M. Efrusy, Olympia Fields Osteopathic Hospital; S. Esposito, Endoscopy Site; J. Fidelholtz; R. D. Folan; R. Frachtman, Center for Clinical Research-Austin; S. Freeman; S. P. Gaddam; G. Gibbons, Remington-Davis, Inc; J. M. Gordon, University Hospitals of Cleveland; D. E. Gremillion, Nashville Research Associates; A. Harris, New York Hospital Medical Center of Queens; M. M. Jamal, UNM Health Sciences Center; D. S. James; J. F. Johanson, Rockford Gastroenterology Association; D. A. Johnson, Digestive & Liver Disease Specialists; J. Jolley; L. Y. Korman, Metropolitan Gastroenterology Group, PC; J. A. Ladenheim, Mitchell Pappas Associates; J. Leavitt; L. J. Lifton; M. Madan; H. N. Maimon, Dayton Area Research Association; P. R. McNally, Eisenhower Army Medical Center; J. Medoff; P. J. Milman; V. Motaparthy, Gastroenterology/Endoscopy; A. Rodriguez, Digestive Disease Clinic; W. A. Ross; B. Rubin; S. C. Schindler, Central Kentucky Research Associates, Inc.; H. I. Schwartz, Miami Research Associates; U. Shah, Philp J. Bean Medical Center; M. J. Shaw, Park Nicollet-HQ; H. Siegel, Eastside Comprehensive Medical Center; E. J. Spiotta; W. H. Taub; J. Turse, Health Advance Institute; N. B. Vakil, Sinai Samaritan Medical Center; P. Witt, Greeley Medical Clinic; L. D. Wruble, T. Zarchy; A. M. Zfass, Medical College of VA Hospitals.