This phase 3, randomized, double-blind study was conducted at 23 sites in the United States from August 2004 through March 2006. The protocol and informed consent/assent forms were approved by an Institutional Review Board at each site prior to the start of the study. Children 5 to 17 years of age with the ability to understand the requirements of the study, willingness to comply with all study procedures, documented proctititis, proctosigmoiditis, pancolitis, or indeterminate colitis, and confirmed or suspected active ulcerative colitis were eligible for the study. Patients were diagnosed via colonoscopy within 1 month of screening with mild-to-moderate UC based on a Modified Sutherland UC activity index (MUCAI)19
(). Mild-to-moderate active UC was defined as a total score between 4 and 10, and a score of at least 1 on the rectal bleeding, mucosal bleeding, and physician’s rating of disease activity scale. The FDA agreed that the MUCAI was an acceptable tool to assess the overall disease activity of each patient and efficacy of the study drug during the study.
Components of the Modified Sutherland Ulcerative Colitis Activity Index
Exclusion criteria included: intolerance or allergy to salicylates; previous intolerance to balsalazide or treatment failure on balsalazide; severe UC (MUCAI > 10); significant bowel distention or tenderness associated with guarding or rebound; infectious, ischemic, or immunologic diseases with gastrointestinal involvement; clinically significant hepatic disease (twice the upper normal limit of serum transaminases); clinically significant renal disease (1.5 times the upper normal limit of serum creatinine or blood urea nitrogen); unstable cardiovascular or pulmonary disease; condition or circumstance that would prevent completion of the study or interfere with analysis of study results, including history of drug or alcohol abuse, mental illness, or noncompliance with treatment or visits; pregnancy or breastfeeding; previous treatment in the study; participation in an investigational drug or device study within 30 days prior to study screening; and active malignancy within the last 5 years, except basal cell carcinoma of the skin, or if female, in situ cervical carcinoma that had been surgically excised. Patients were not permitted to take anticholinergic, antidiarrheal, or 5-ASA products once randomized and throughout the study, provided there was no potential risk associated with the subject not receiving these products. Parental or legal guardian consent and patient assent were obtained prior to the start of the study for each patient.
The dosing regimens selected were based on the minimal systemic absorption and local therapeutic action of balsalazide, as well as a survey of prescribing patterns of pediatric gastroenterologists in the US. The survey revealed that the majority of pediatric gastroenterologists routinely treated their patients with mild-to-moderate active UC with balsalazide 6.75 g/day and then tapered the dose for maintenance. The lower dose was selected for this study because it was associated with improved endoscopic scores and physician’s global assessment of disease when administered to adults with mild-to-moderate active UC.18
A placebo was not selected because of ethical issues associated with using a placebo in pediatric patients with UC.
Patients were randomized in a 1:1 ratio using a computer-generated randomization schedule to receive balsalazide 6.75 g/day or 2.25 g/day. Randomization was stratified into 3 of the following groups based upon age and agreement to participate in the pharmacokinetic (PK) analysis: patients 5 to 8 years of age; patients 9 to 17 years of age and participating in the PK portion of the study; and patients 9 to 17 years of age and not participating in the PK portion of the study. Patients were provided 8 weeks of study medication packaged in 2-week supply cartons. Daily doses, consisting of 3 tablets each, were taken orally 3 times a day for a total of 9 tablets daily, approximately 8 hours a part, for a total of 8 weeks. If patients were unable to swallow the capsules, the parents were instructed to open the capsules and sprinkle the contents on food for immediate consumption. All capsules were identical in appearance and contained either 750 mg of balsalazide or the inert carrier alone. Both investigator and patient were blinded to the study drug assignment throughout the course of the study. Patients were instructed to return used and unused study medication during the 2-week, 4-week, and 8-week follow-up visits to determine compliance with study medication. Patients, parents, or patient guardians were contacted weekly between study visits by phone to assess patient symptoms and to determine compliance.
The following items were obtained prior to randomization: medical history, physical examination, vital signs, urine pregnancy test (for females), record of concomitant medications, routine laboratory studies (chemistry, hematology, serum transaminases, urinalysis, and creatinine clearance test), baseline MUCAI score, and colonoscopy with biopsies. Patients who had a colonoscopy and biopsy within 1 month of the screening visit did not have the procedure repeated provided that adequate documentation diagnosing UC was on file. Patients received diary cards to record the presence of abdominal cramps and fever ≥ 37.8C between visits.
Patients returned for follow-up visits at 2 weeks (± 3 days), 4 weeks (± 3 days), and 8 weeks (± 3 days). The following were performed or obtained at the 2-week and 4-week visit: vital signs, urine pregnancy test (for females), record of concomitant medications, record of adverse events and serum creatinine. Diary cards were also collected during these visits. Plasma concentration measurements for balsalazide, 5-aminosalicylic acid (5-ASA), and N-acetyly-5-aminosalicylic acid (N-Ac-5-ASA) were performed at the 2-week visit. Maximum observed plasma concentration (Cmax), area under the plasma concentration-time curve (AUC), and time to maximum plasma concentration (Tmax) values were determined.
The following were performed or obtained during the 8-week visit: physical examination, vital signs, urine pregnancy test (for females), record of concomitant medications, record of adverse events, routine laboratory studies (chemistry, hematology, serum transaminases, urinalysis, and creatinine clearance test), diary card, MUCAI score, and colonoscopy with biopsy. Three to four colonic biopsies were obtained from the same general regions that were biopsied during the screening period. Histologic assessment was performed and graded on a scale from 0 to 3 (0: no cryptitis or crypt abscesses; 1: cryptitis or abscesses involving < 50% of sampled crypts; 2: cryptitis or abscesses involving ≥ 50% of sampled crypts; 3: active inflammation plus erosions or ulcerations).
The study subjects could discontinue from the study at any time but were withdrawn if they experienced an increase in stool frequency for 3 consecutive days defined as 2 or more additional stools per day or a clinically relevant increase in rectal bleeding based on the investigator’s assessment of patient diary responses. Regardless of the reason of the withdrawal, all patients were asked to undergo an end-of-therapy evaluation (Week 8 visit).
The primary efficacy endpoint was the proportion of all randomized patients [intention-to-treat (ITT) population] achieving clinical improvement, which was defined as a reduction from baseline in the MUCAI total score by at least 3 points at week 8. A Fisher’s exact test was used to determine any significant difference between the treatment groups. An exact (Clopper-Pearson) two-sided 95% confidence interval to determine the proportion of subjects with clinical improvement was calculated. The same analysis was performed on the protocol population which included patients who did not violate the protocol in any fundamental way and who were at least 70% compliant in taking study medication. An analysis of factors influencing the primary endpoint of clinical improvement was performed in the ITT population. A logistic regression model (Wald chi-sqare test) was used for this statistical analysis with a 2-sided significance level of 0.05. Variables in the initial model included treatment, MUCAI total score at baseline, age, sex, time since colonoscopy with biopsies, and duration of ulcerative colitis.
Secondary measures of efficacy, which was based on the ITT population, included: the proportion of patients achieving remission, as evidenced by a score of 0 or 1 (a score of 1 was only allowed on the stool frequency index) on the MUCAI at week 8; change from baseline to week 8 in the total MUCAI score defined as the individual items (stool frequency, rectal bleeding, mucosal appearance, physician’s rating of disease activity) of the MUCAI, and the pathology classification of the colonic biopsies; and number of days abdominal cramps and/or fever were reported on the individual patient diary card assessment in the 7 days prior to the 4-week and 8-week visits. The following scale was used to assess the categorical change in stool frequency, rectal bleeding, mucosal appearance, physician’s rating of disease activity, and pathology classification:
- improved = reduction from baseline to week 8 of MUCAI subscore and pathology classification
- unchanged = MUCAI subscore was the same at baseline and at week 8
- worsened = increase from baseline to week 8 of the MUCAI subscore.
A Fisher’s exact test was used to determine any significant difference between treatment groups in achieving remission at week 8. A p-value and an approximate 2-sided 95% confidence interval based on the normal distribution for the difference in proportion achieving remission between the 2 groups were calculated. An analysis of covariance (ANCOVA) with treatment group as a factor and baseline value as a covariate was used to evaluate the change from baseline to week 8 in the total MUCAI score. A Wilcoxon rank-sum test was used to evaluate the following secondary endpoints: change from baseline to week 8 in the individual items (stool frequency, rectal bleeding, mucosal appearance, physician’s rating of disease activity) of MUCAI, change from baseline in pathology classification at week 8, and number of days that abdominal cramps and/or fever were reported on individual subject diary cards in the 7 days prior to the 4-week and 8-week visit.
All statistical tests were 2-tailed at the 5% level of significance. Missing data were entered for the ITT population using the last-observation-carried-forward method for the primary and secondary endpoints. Baseline observations were not carried forward if all subsequent measurements were missing.
Pharmacokinetic (PK) analyses were performed for the PK population on plasma concentration-time data, and computed parameters were calculated from this data. The sample size for the study was based on what was considered feasible to obtain rather than standard considerations of statistical power.