Adrenocorticotropic Hormone (ACTH) for Frequently Relapsing and Steroid Dependent Nephrotic Syndrome

Our hypotheses are the following: Hypothesis 1: ACTH gel is superior to no treatment in maintaining remission in children with frequently relapsing or steroid dependent nephrotic syndrome (NS). Hypothesis 2: Relapses in children with frequently relapsing or steroid dependent nephrotic syndrome receiving ACTH gel will increase when the dose of ACTH gel is reduced by 50%. Hypothesis 3: ACTH gel will increase the percentage of children with frequently relapsing or steroid dependent nephrotic syndrome that remain relapse free off medication. Primary end-points: The primary end-point related to Hypothesis 1 is the proportion of patients in each arm with a relapse during the initial 6 months of treatment. The primary end-point related to Hypothesis 2 is the proportion of relapse-free patients during the first 6 months and second 6 months of treatment with ACTH. This will include patients initially randomized to ACTH and patients who receive ACTH as rescue therapy following their initial relapse in patients randomized to no treatment. The primary end-point related to Hypothesis 3 is the proportion of patients who have relapses in the 6 months following completion of one year of ACTH. This will be compared to the proportion of patients with relapses during the initial 6 months in the patients randomized to no treatment. Secondary end-points: Our secondary end-points are the following: 1. The total prednisone exposure over the initial 12 months in the two groups. 2. The number of relapses over the initial 6 months in the two groups. 3. The change in body mass index (BMI), height standard deviation score (SDS), and cholesterol over the study period for both groups Study Design and Methods The experimental design is a multi-center, prospective, controlled open label, randomized trial comparing ACTH gel and no treatment in preventing relapses in pediatric patients with frequently relapsing or steroid dependent nephrotic syndrome. Patients will be randomized in a 1:1 ratio to either no treatment or treatment with ACTH gel. The primary outcome will be the presence of a relapse within 6 months of starting ACTH gel or no treatment. After initial recruitment, enrollment will begin with a screening visit to determine eligibility and obtain informed consent and assent. Randomization and weaning of all other medications for the treatment of NS will begin after remission has been achieved for those with active relapse. There will be a 2 week overlap of ACTH and current immunosuppressive medications: 1. ACTH will be initiated 2 weeks prior to the completion of the prednisone taper for patients who are receiving prednisone at the time of consent. 2. ACTH will be initiated 2 weeks prior to stopping preventive medications such as tacrolimus, cyclosporin, and mycophenolate. Patients randomized to no treatment will be followed for up to 6 months or until disease relapse, whichever occurs first. Patients who relapse within 6 months will be given the option of reassignment to the ACTH treatment group after remission has been achieved using conventional corticosteroid therapy. Patients randomized to ACTH treatment will be given ACTH for 12 months. During the second 6 months, the ACTH dose will be reduced to 50% of the starting dose. The outcome of interest is the presence of relapses after dose reduction. Follow-up will occur throughout the 12 months of therapy, and also for 6 months following the completion of ACTH therapy. The outcome of interest is the percentage of patients with relapses in the 6 months after completing a 12 month course of ACTH treatment. The primary end-point of the study and on which the statistical power is based is the proportion of patients who have a relapse in the 6 months following randomization to either ACTH or no ACTH. We hypothesize that the 6 month relapse rate for patients receiving no treatment is 70%. In order to detect a 6 month relapse rate of 30% for patients receiving the ACTH gel, we will randomize 30 patients in each arm using a two sided z test with alpha=.05. Our statistical power to detect such a difference is 91% which assumes two interim analyses at 50% and 100% of accumulated information. That is, once 30 patients have 6 month relapse data, we will conduct the first interim analysis. The last one will be completed once we have relapse data on all 60 patients. The table below gives the operating characteristics for such a design: Number of Patients with 6 Months Data Boundary p-value 30 0.006 60 0.045 Using the method of Lan-DeMets, we list the boundary p-value for this sequential design. Thus, after our first interim look, we will reject the null hypothesis of equal 6 month relapse rates between ACTH gel and no treatment if our test statistic renders a p-value \< .006. According to the intent to treat principle, patients will be analyzed according to the treatment they have been assigned to during the randomization procedure. The odds ratio of ACTH versus no ACTH, plus the Wald 95% confidence interval, will be also be calculated. The primary end-point related to specific aim 2 is the proportion of relapse-free patients during the first 6 months and second 6 months of treatment with ACTH. This will include patients initially randomized to ACTH and patients who receive ACTH as rescue therapy following their initial relapse. We will estimate 6 month and 12 month relapse-free rate using the method of Kaplan-Meier and compare treatments using a log-rank test. The primary end-point related to specific aim 3 is the proportion of patients who have relapses in the 6 months following completion of one year of ACTH. This will be compared to the proportion of patients with relapses during the initial 6 months in the patients randomized to no treatment using a z test statistic. Odds ratios of ACTH vs no ACTH will also be calculated. We will also compare patients as randomized by secondary endpoints such as total prednisone exposure in 12 months, number of relapses, cholesterol and change in BMI with two sample t-tests. If normality assumptions do not hold, appropriate non-parametric methods will be used. Growth data collected during the study will be summarized descriptively for each treatment group at each time point. Based on height data collected during the study and published reference height information, the height standard deviation score (SDS, also called z-score) will be computed for each patient at each time point as: (Height - mean height for that age category) / SD of height for that age category. Descriptive statistics of this endpoint will be presented by time point and the z-scores will allow identification of potential outliers. Treatment Assignment and Randomization Treatment assignments will be stratified according to clinical center. The treatment assignments will be generated by the Data Coordinating Center (DCC) with the use of a pseudo-random-number generator with randomly permutated blocks that will be used to ensure balance between the number of subjects assigned to each treatment (ACTH or no ACTH). Before the study starts, the institutional research coordinator at each clinical center will be given a batch of 20 sealed, sequenced, opaque envelopes containing the treatment assignment and will have a unique identification number consisting of the clinical center stratum. Patients will be assigned to one of the two treatment arms in a ratio of 1:1.