An Oral Preparation of Mesalamine as Long-Term Maintenance Therapy for Ulcerative Colitis

Methods

The study protocol was approved by the institutional review board of each study site, and each patient gave written, informed consent before study entry.

Patients

We recruited patients from 17 study sites: 8 private practices, 5 university-based medical centers, and 4 hospitals or clinics. Recruiting was done using print and radio advertising, posters, and announcement letters to referring physicians and local support groups. Each patient enrolled in the trial was between 18 and 75 years of age and had a documented diagnosis of ulcerative colitis. At the time the study began, all patients had been in remission for at least 1 month as indicated by the endoscopic appearance of the bowel and by the passage of five or fewer bloodless stools per day. To qualify for study entry, patients were required to have a score of 0 on the proctosigmoidoscopic grading scale described below (see Evaluation Methods). The presence of colitis symptoms, such as loose stools or abdominal cramps, was not a reason for exclusion from the study, provided that endoscopic examination showed remission of disease. All patients had been treated previously with 2.0 to 4.0 g of sulfasalazine per day or 0.8 to 1.6 g of any oral mesalamine product per day; the dosage of these drugs had been constant for at least 1 month before study entry. No patient had received corticosteroid or topical rectal therapy within 1 month of study entry.

Female patients with child-bearing potential were required to practice a reliable method of birth control throughout the study, and women who were pregnant or nursing were excluded from the study. Patients with a history of allergy or intolerance to aspirin or salicylates were also excluded, as were patients with a history of extensive bowel resection causing the short-bowel syndrome and patients with laboratory evidence of renal or hepatic dysfunction.

Throughout the study, patients were not allowed to use corticosteroids (except topically for dermatologic reasons), sulfasalazine, antibiotics for more than 10 consecutive days, topical rectal therapies, or investigational drugs other than mesalamine.

Study Design

In this multicenter, double-blind study, 264 patients were randomly assigned to receive one of three treatment regimens: placebo, 0.8 g of oral mesalamine per day, or 1.6 g of oral mesalamine per day. Randomization was done within centers by means of randomization codes using specific patient numbers generated for each study site before the study began. Each patient number was randomly distributed by computer to one of the three treatment groups. Patients were not stratified according to any clinical characteristic, such as disease location or previous treatment.

Patients were seen at baseline and at months 1, 3, and 6 of treatment. They were also seen at any time during the study if they felt that their disease was worsening or that they were having an adverse event. At baseline, each patient had a screening at which medical history was recorded and physical examination, proctosigmoidoscopy or colonoscopy, laboratory analyses, and pregnancy testing were done. At each follow-up visit, physical examination, proctosigmoidoscopy or colonoscopy, laboratory analyses, and pregnancy testing were done. Throughout the study, patients maintained a daily diary in which they recorded their study medication dosing, their symptoms, and their concomitant medication dosing.

Study Drug

The mesalamine used in our study (Asacol) was supplied in tablets that consisted of a 400-mg mesalamine core enveloped by a pH-sensitive acrylic polymer coating (Eudragit-S, Rohm Pharma GmbH, Weiterstadt, West Germany). Because the pH-sensitive resin breaks down only at a pH of 7 or more, the coating remains intact until the tablet reaches the terminal ileum, where the pH is consistently higher than 7. Radiologic studies have shown that barium-containing capsules coated with this resin break down in the terminal ileum and colon [22, 23]; thus, this resin makes it possible to provide topical application of mesalamine directly to these areas [22]. The placebo tablets were identical to the mesalamine tablets in odor and appearance but contained no active ingredients.

Placebo was administered as four tablets daily. Mesalamine, 0.8 g/d, was administered as two active and two placebo tablets daily, and mesalamine, 1.6 g/d, was administered as four active tablets daily. Tablets were packaged in four bottles, each of which was labeled with the time of day at which patients were to take the tablets in that bottle. All patients took one tablet at breakfast, one at lunch, one at dinner, and one at bedtime. Patients in the mesalamine, 0.8 g/d group took the active tablets at breakfast and at bedtime. The study treatment period was scheduled to last 6 months.

Compliance was monitored by tablet count and by review of patient diaries at each study visit. Noncompliance was defined as missing more than 15% of the study medication over the duration of treatment or more than 50% of the study medication for 4 consecutive days (for reasons other than intolerance).

Evaluation Methods

All patients were evaluated using proctosigmoidoscopy or colonoscopy at baseline and at each visit. Findings were graded according to the following scale: 0 (normal or mild granularity, edema, hyperemia, or erythema; mildly diminished vascular markings); 1 (mild granularity, edema, hyperemia, or erythema; mildly diminished vascular markings plus friability); 2 (marked erythema or granularity; no vascular markings; bleeding with minimal trauma; no ulcerations); and 3 (spontaneous bleeding; ulcerations). A score of 0 was required for study entry. If a patient was noted at baseline to have a score of 0 with mild granularity, edema, hyperemia, or erythema or mildly diminished vascular markings, the investigator specifically commented on these factors and clearly stated that the patient was considered to be in endoscopic remission. The investigators agreed that these findings could be present in patients with longstanding ulcerative colitis in remission. Subsequently, a score of 1 or higher at any time was considered indicative of relapse.

Safety was evaluated at each visit using physical examination and laboratory analyses. Laboratory analyses consisted of routine biochemical and hematologic measurements and urinalyses done at each visit, plus creatinine clearance assays to monitor renal function done at the baseline visit and the final visit. Patients were questioned at each visit about any adverse events they may have had. In addition, any events noted in patients diaries that were considered to be possible adverse events were tabulated as adverse events.

Efficacy Variables

The primary efficacy variable was treatment outcome. Treatment outcome could be either success, defined as maintenance of remission (as indicated by endoscopic evaluation) at the 6-month study visit, or failure, defined as endoscopic relapse at any time during the study or withdrawal due to an adverse event. Time to relapse was also evaluated.

Statistical Analysis

The primary efficacy analysis included patients who were compliant with the protocol until they had completed 6 months of therapy or until they had discontinued treatment because of relapse or adverse events. In addition, an intention-to-treat analysis, which included all patients exposed to treatment, was also done. In this analysis, treatment success was determined on the basis of remission at the patient's final visit, regardless of whether the patient had completed 6 months of treatment.

Before the study began, a sample size of 64 patients per treatment group was determined to be sufficient to detect a 25% difference in proportions of patients having relapse, using a two-sided 0.05 significance level with 0.80 power. Because more patients than expected could not be analyzed, the number of patients per treatment group was increased during the course of the study before the study was unblinded.

The baseline characteristics of patients in the two mesalamine groups were compared with those of patients in the placebo group by using the t-test for continuous variables and the Fisher exact test for categorical variables.

The treatment outcome of patients in the two mesalamine groups was compared with the treatment outcome of patients in the placebo group by using the Fisher exact test. Treatment outcome was also analyzed by treatment group and age, sex, and race by using categorical methods, log-linear model.

Time to relapse in the mesalamine groups was compared with time to relapse in the placebo group by survival analysis using the product-limit method to compute estimates of the survival function (remission) and using the log-rank test to examine treatment group differences. In this analysis, patients who did not have relapse were considered to be censored at the last date of study participation, and patients in whom treatment was discontinued prematurely because of an adverse event were considered to be censored at the date of discontinuation.

The distribution of the number of adverse events recorded per patient in the mesalamine groups was compared with that in the placebo group by contingency Table analysis using the Pearson chi-square statistic with the number of adverse events per patient categorized as 0, 1 to 5, 6 to 10, or more than 10.

Results

Patients

Two hundred sixty-four patients were randomly assigned to receive either placebo, 0.8 g of mesalamine per day, or 1.6 g of mesalamine per day. Randomization was done at all 17 study sites. The number of patients at each site ranged from 1 to 47 (mean, 16 patients). The demographic and disease history characteristics of the 264 patients are presented in Table 1. The characteristics of the mesalamine groups were similar to those of the placebo group, except for the sex distribution: The proportion of women in the mesalamine, 1.6 g/d group (57.5%) was greater than that in the placebo group (38.9%) (P = 0.015). In general, there was good agreement between clinical and endoscopic remission at baseline.

Seventy-five patients were excluded from the primary efficacy analysis for the following reasons: failure to meet study entry criteria (n = 36), noncompliance with study medication (n = 18), noncompliance with study procedures (n = 3), concomitant medication violation (n = 10), loss to follow-up (n = 4), and voluntary withdrawal (n = 4). The numbers of patients excluded were similar in the three groups.

Treatment Outcome

Analyses of data from patients who completed 6 months of treatment or who withdrew because of relapse or adverse events (primary efficacy analyses) showed that the criteria for treatment success were met in 25 of 63 patients (39.7%) in the placebo group compared with 40 of 68 patients (58.8% [95% CI, 46.4% to 71.3%]) in the group receiving mesalamine, 0.8 g/d, and 38 of 58 patients (65.5% [CI, 52.4% to 78.6%]) in the group receiving mesalamine, 1.6 g/d (Table 2). The percentage of patients who met the criteria for treatment success was greater in both the group receiving mesalamine, 0.8 g/d (P = 0.036) and the group receiving mesalamine, 1.6 g/d (P = 0.006) than in the placebo group. In the intention-to-treat analysis of all patients, the differences between the placebo and the mesalamine groups were also significant: The criteria for treatment success were met in 42 of 87 patients (48.3%) in the placebo group compared with 57 of 90 patients (63.3% [CI, 52.8% to 73.8%]) in the group receiving mesalamine, 0.8 g/d (P = 0.050) and 61 of 87 patients (70.1% [CI, 59.9% to 80.3%]) in the group receiving mesalamine, 1.6 g/d (P = 0.005). Subgroup analyses of treatment outcome by age, sex, and race did not show any association between these baseline factors and treatment outcome (P > 0.05). Throughout the study, agreement was generally good between symptomatic and endoscopic relapse in all groups. Treatment was discontinued in two patients, one in the placebo group and one in the group receiving mesalamine, 1.6 g/d, because of the appearance of symptoms of ulcerative colitis, despite normal proctosigmoidoscopic findings.

Time to Relapse

Table 3 and Figure 1 show that in the primary efficacy analysis of compliant patients who completed 6 months of therapy or who withdrew because of relapse or adverse events, both of the mesalamine groups differed from the placebo group with respect to time to relapse (P = 0.026 for the group receiving mesalamine, 0.8 g/d; P = 0.011 for the group receiving mesalamine, 1.6 g/d). The survival plots of the patients who received placebo and the patients who received mesalamine diverge after approximately 30 days; patients who received mesalamine showed a greater probability of maintaining remission throughout the rest of the study. In the intention-to-treat analysis of all patients, a statistically significant difference in survival distribution was seen between the group receiving mesalamine, 1.6 g/d and the placebo group (P = 0.008) but not between the group receiving mesalamine, 0.8 g/d and the placebo group (P = 0.074).

In a supplemental analysis, in which patients were stratified according to extent of disease (pancolitis, left-sided disease, proctosigmoiditis, proctitis, and unspecified), the distributions of time to relapse were similar in the five groups (P = 0.907).

Compliance

Six of 87 patients (6.9%) in the placebo group, 11 of 90 patients (12.2%) in the group receiving mesalamine, 0.8 g/d, and 4 of 87 patients (4.6%) in the group receiving mesalamine, 1.6 g/d, met the criteria for study drug noncompliance. All 21 of these patients were dropped from the study, some because of dosing noncompliance alone and others for this reason plus others.

Adverse Reactions to Therapy

The number of investigator-recorded adverse events and the percentage of patients who reported adverse events were similar among the three treatment groups. Thirty-four of 87 patients (39.1%) in the placebo group reported 81 events; 29 of 90 patients (32.2%) in the group receiving mesalamine, 0.8 g/d, reported 72 events; and 36 of 87 patients (41.4%) in the group receiving mesalamine, 1.6 g/d reported 106 events. The adverse events recorded most frequently by the investigators were headache, flu syndrome, diarrhea, rhinitis, and abdominal pain. When potential adverse events noted in patient diaries were combined with investigator-recorded adverse events, statistical analysis of the distribution of the number of adverse events reported per person (0, 1 to 5, 6 to 10, or more than 10) showed no significant differences between the placebo group and the group receiving mesalamine, 0.8 g/d (P = 0.922) or between the placebo group and the group receiving mesalamine, 1.6 g/d (P = 0.236).

Most adverse events were mild to moderate in severity and were not serious. Three patients, one from each treatment group, had serious adverse events: One patient in the placebo group had chest pain, hypertension, and dyspnea that were considered to be unrelated to treatment; one patient receiving mesalamine, 0.8 g/d, had a questionable transient ischemic attack and migraine, the relation of which to treatment was considered doubtful; and one patient receiving mesalamine, 1.6 g/d, had a miscarriage that was considered to be unrelated to treatment. Treatment was discontinued because of an adverse event in 4 of 87 patients (4.6%) in the placebo group, 4 of 90 patients (4.4%) in the group receiving mesalamine, 0.8 g/d, and 2 of 87 patients (2.3%) in the group receiving mesalamine, 1.6 g/d. The events leading to discontinuation were different for each patient, with the exception of headache and paresthesia. Headache was the reason for discontinuation in one patient in the placebo group and one patient in the group receiving mesalamine, 0.8 g/d, and paresthesia was the reason for discontinuation in two patients in the placebo group. None of these adverse events was considered serious.

We detected no clinically significant trends in any of the laboratory test results, no treatment-related abnormalities in the results of biochemical or hematologic tests or urinalyses, and no changes in creatinine clearance. No patient was dropped from the study prematurely because of laboratory abnormalities.

Discussion

In our study, daily dosages of 0.8 and 1.6 g of the pH-sensitive, polymer-coated oral mesalamine preparation were safe and effective in maintaining remission in patients with quiescent ulcerative colitis. After 6 months of treatment, 58.8% of the patients receiving mesalamine, 0.8 g/d, and 65.5% of the patients receiving mesalamine, 1.6 g/d remained in remission, compared with 39.7% of patients receiving placebo. Sex, age, and race were not found to predict treatment success or failure.

Relapse rates for patients with quiescent ulcerative colitis may depend on the length of follow-up and, probably, on the dose of the maintenance drug, although further studies are indicated. In addition, comparisons of the results of clinical trials of maintenance treatments for ulcerative colitis are confounded by the different definitions of relapse and the different measures of treatment success used in the various studies. Recognizing that these factors make direct comparisons between studies problematic, we present relapse rates reported in other comparative trials of sulfasalazine and mesalamine products (see Table 4) to provide a context for our findings. In our study, the rates of relapse over 6 months (excluding patients who were withdrawn because of adverse events) were 35.3% for patients receiving 0.8 g of mesalamine per day, 31.0% for patients receiving 1.6 g of mesalamine per day, and 54.0% for patients receiving placebo. These rates were within the range of those reported by other investigators.

We used the proctosigmoidoscopic appearance of the bowel as the indicator of whether relapse had occurred. Because of the spectrum of normal bowel habits and their overlap with “irritable bowel” symptoms in patients with ulcerative colitis and with occasional blood from benign anal conditions, we believe that proctosigmoidoscopy is the most objective way to determine the level of disease activity. Proctosigmoidoscopic findings generally correlate well with the level of symptoms the patient is having, but we used them to clarify discrepancies between subjective symptoms (bowel frequency or liquidity) and to assess hematochezia not related to active colitis. Clearly, an effective maintenance therapy for ulcerative colitis should control symptoms in addition to maintaining the integrity of the bowel mucosa. Our study design did not include the analysis of ulcerative colitis symptoms over the course of treatment. During the study, only two patients were considered to have had a symptomatic relapse not accompanied by proctosigmoidoscopic changes. This provides reassurance that proctosigmoidoscopic remission was accompanied by control of symptoms in almost all cases.

The dropout rate in our study was high, largely because patients failed to meet strict entry criteria. When the study began, patients who had been entered were dropped if their baseline proctosigmoidoscopic examinations showed even mild granularity or erythema or mildly diminished vascular markings. As a result, many patients considered to be in clinical remission were excluded from the study on the basis of minor proctosigmoidoscopic findings. During the course of the study, the proctosigmoidoscopic grading scale was changed to allow entry of patients with these mild findings, because the investigators agreed that patients with longstanding ulcerative colitis in remission may have had mild granularity, edema, hyperemia, or erythema or mildly diminished vascular markings on endoscopy.

The polymer-coated oral mesalamine preparation was well tolerated; most adverse events were mild or moderate in severity. The percentages of patients having adverse events and the number of adverse events that patients had were similar in the placebo group and the mesalamine groups. This suggests that the adverse events reported may have been related to the symptoms of ulcerative colitis rather than to any effects of the active treatment. Furthermore, no clinically meaningful dose-response relation was apparent for any of the adverse events reported. The adverse events seen in our study were similar to those reported in other studies of mesalamine as maintenance therapy for ulcerative colitis [19, 21] and to those most frequently reported by patients receiving sulfasalazine [29]. In previous studies done in rats [30], the intravenous administration of mesalamine was associated with renal toxicity. Careful monitoring of renal function in our patients showed no clinically significant untoward changes in any of the renal function variables studied. This finding is consistent with the findings of other clinical trials of treatment of ulcerative colitis with mesalamine [15, 16, 21].

Our results indicate that coated mesalamine at dosages of 0.8 g/d and 1.6 g/d is safe and effective in maintaining remission in patients with quiescent ulcerative colitis.

Appendix

The following are members of The Mesalamine Study Group: Stephen B. Hanauer, MD, University of Chicago Hospital, Chicago, Illinois; Charles A. Sninsky, MD, University of Florida, Gainesville, Florida; Malcolm Robinson, MD, University of Oklahoma College of Medicine, Oklahoma City, Oklahoma; Bernard J. Powers, MD, Arapahoe Gastroenterology, Englewood, Colorado; James D. McHattie, MD, Pasqua Hospital, Regina, Saskatchewan, Canada; James E. Mayle, MD, Michigan State University, Lansing, Michigan; Charles O. Elson, MD, University of Alabama at Birmingham, Birmingham, Alabama; Michael P. DeMicco, MD, Associated Gastroenterology Medical Group, Anaheim, California; James H. Butt, MD, University of Missouri School of Medicine, Columbia, Missouri; Ronald E. Pruitt, MD, Nashville Medical Research Institute, Nashville, Tennessee; John M. Bozdech, MD, The Cleveland Clinic, Cleveland, Ohio; Michael A. Safdi, MD, Greater Cincinnati Gastroenterological Associates, Cincinnati, Ohio; Michael S. Gurney, MD, The Portland Clinic, Portland, Oregon; Alan M. Fixelle, MD, Atlanta, Georgia; Arnold I. Levin, MD, Eastside Digestive Disease Clinic, Kirkland, Washington; John Smoots, MD, The Western Clinic, Tacoma, Washington; and Douglas C. Wolf, MD, Emory University, Atlanta, Georgia.

Dr. Sninsky: Division of Gastroenterology, Gainesville Veterans Administration Center, Section 111 C, Room E 330, 1600 Archer Road, Gainesville, FL 32608.

Dr. Robinson: Oklahoma Foundation for Digestive Research, 711 SL Young Boulevard, Suite 501, Oklahoma City, OK 73104.

Dr. Powers: Arapahoe Gastroenterology, 950 East Harvard, Suite 540, Englewood, CO 80210.

Dr. McHattie: Gastrointestinal Unit, Pasqua Hospital, 4101 Dewdney Avenue, Regina, Saskatchewan, S4T 1A5, Canada.

Dr. Mayle: Michigan State University Gastroenterological Services, Ingham Medical Professional Center, 405 West Greenlawn Avenue, Suite 130, Lansing, MI 48910.

Dr. Elson: Division of Gastroenterology, University of Alabama at Birmingham, 701 South 19th Street, Birmingham, AL 35294.

Dr. DeMicco: Associated Gastroenterology Medical Group, 1222 West La Palma, Suite 306, Anaheim, CA 92801.

Dr. Butt: University of Missouri School of Medicine, Truman Veterans Administration Hospital, 800 Hospital Drive, Columbia, MO 65201.

Dr. Pruitt: Nashville Medical Research Institute, 4230 Harding Road, Suite 309, Nashville, TN 37205.

Dr. Bozdech: QPS Clinic, 14th and Main Streets, Quincy, IL 62301.

Dr. Safdi: Greater Cincinnati Gastroenterological Associates, 2925 Vernon Place, Suite 100, Cincinnati, OH 45219.

Dr. Gurney: West Hills Gastroenterology Association, 9155 South West Barnes Road, Suite 636, Portland, OR 97225-6652.

Dr. Fixelle: 4470 North Shallowford Road, Suite 203, Atlanta, GA 30338.

Dr. Levin: Eastside Digestive Disease Clinic, 13030 121st Way NE, Suite 201, Kirkland, WA 98034.

Dr. Smoots: Franciscan Family Care, 1708 South Yakima, Tacoma, WA 98405.

Dr. Wolf: 5671 Peachtree-Dunwoody Road, Suite 550, Atlanta, GA 30342.