Determination of Sample Size. The sample size is based on a fixed-sequence strategy which tests for a treatment difference in the primary endpoint (percent change at Week 12 in liver fat from baseline as determined by MRI-PDFF) between each randomized TVB-2640 dose group (25 and 50 mg) versus placebo. The fixed-sequence strategy will start with the highest dose group comparison. If statistically significant, then testing will proceed to the low dose group comparison. Each TVB-2640 dose group will be compared to the pooled placebo group using an F-test test from an analysis of covariance (ANCOVA) model with fixed effects for the stratification factor (diabetes presence/absence) and treatment group (i.e., TVB-2640 dose groups and pooled placebo) and with the baseline MRI-PDFF value as a covariate. Since the primary efficacy endpoint may not be normally distributed, the non-parametric Wilcoxon rank-sum test is conservatively used, instead of the F-test, for power calculations. The sample size is based on a conservative assumption that the primary analysis is going to be performed in the US population only, resulting in 30 evaluable subjects in the placebo group, and 30 evaluable subjects in each of the randomized TVB-2640 dose levels. Based on Xxxxx (2016), it is conservatively assumed that the primary endpoint has a standard deviation of 30. Power calculations assume the primary endpoint is lognormally distributed with a standard deviation of 30, there are at least 30 evaluable subjects in each treatment group, and the two-sided Wilcoxon rank sum test will be used to test each pairwise treatment difference at the 0.05 Type I error level. Under the fixed-sequence strategy which maintains an overall 0.05 Type I error rate, the study has at least 80% overall power to detect both treatment differences (i.e., both high dose [50 mg] versus placebo and low dose [25 mg] versus placebo), if each mean treatment difference is at least 24. If the study does not proceed to the high-dose level, (i.e., there are 30 evaluable subjects treated with 25 mg TVB-2640 and only 15 with placebo), then the study has at least 77% power to detect a mean treatment difference of at least 24. The study will have more power if the subjects from China are included in the primary analysis and the same assumptions of treatment difference and variability hold. Power calculations for the key secondary endpoint (percentage of subjects with at least a 30% reduction in the primary endpoint) are based on pairwise comparisons w...
Determination of Sample Size. It is expected that the responder rate for this study is approximately 80% based on the results of a pivotal study of a similar investigational product Juvéderm Volbella XC. The margin of error (associated with 95% CI) for estimating an 80% expected responder rate is 10%, which is desirable to attain by having a sample size of 60 subjects for the study. Patients’ dropout rate is expected to be negligibly small at Day 30 post-treatment when the primary efficacy assessment is collected, thus 60 subjects will be recruited into the study.
Determination of Sample Size. The sample size of this study is based on the results obtained with a randomized, double blind, study with Racecadotril in the treatment of hospitalized children aged 3-60 months suffering from acute watery diarrhea (Bio-Projet Study nr. 45, Xxxxxxx Xxxxx et al. 1998). A total of 135 subjects were analyzed in this study. Data of the study showed following recovery rates over time: Duration of treatment Percentage of recovered patients Placebo Racecadotril Sample size per group required for a power of 80% using log-rank test (+10% drop out rate) after 1 day (24 hours) 18.3% 46.6% 41 (46) After 2 days (50 hours) 35.8% 70.1% 36 (40) After 3 days (72 hours) 59.4% 84.2% 55 (62) After 4 days (96 hours) 76.3% 90.1% 119 (132) After 5 days (120 hours) 80.6% 94.7% 90 (100) The mean duration of diarrhea was 64 (±4.6) hours in the placebo group (median 64 hours) and 40 (±4.1) hours in the Racecadotril group (median 28 hours), Therefore an appropriate approach to investigate a difference between recovery rates of the two treatment groups would be after a treatment duration of at least 3 days (i.e 72 hours).When the sample size in each group is 55 (with a total number of recoveries of at least 26), the two-sided log-rank test for equality of survival curves will have 80% power to detect the difference between 84.2% rate of recovery in the Racecadotril group and the 59.4% rate of recovery in the placebo group after a treatment duration of 72 hours. Adding a 10% drop-out rate the sample size per groups would be 62 subjects i.e. 124 subjects in total. A minimum number of 20 subjects should be available in each age group (3 – < 24 months, 2 years to 11 years, 12 to < 18 years). A subgroup analysis per age group (3 – < 24 months, 2 years to 11 years, 12 to < 18 years) will be performed.
Determination of Sample Size. There is no formal sample size calculation. The sample size is based on qualitative considerations and is sufficient to provide safety evaluation of eteplirsen in the studied population.
Determination of Sample Size. This is a double-blind study. Each cohort will be independently analyzed. A total of approximately 26 subjects will be randomized in each cohort. The first 12 subjects will be randomized to the KPL-716 and the placebo arms in a 3:1 randomization ratio. The rest of the 14 subjects will be randomized in a 1:1 randomization ratio. There will be approximately 16 subjects and 10 subjects randomized to the KPL-716 arm and the placebo arm respectively. Based on a two- sample t-test for the primary efficacy endpoint, change from baseline in weekly average of WI- NRS at Week 8, assumed mean changes of 4 for the KPL-716 arm and 1.5 for the placebo arm, a total sample size of 26 subjects per cohort will provide about 80% power to detect a 2.5-point mean difference with a standard deviation of 2.8, given a two-sided alpha of 0.2. Given the exploratory nature of this pilot study, the precision attained with the above cohort samples size is acceptable for early signal of efficacy assessments.
Determination of Sample Size. This is a double-blind study. Each cohort will be independently analyzed. A total of approximately 26 subjects will be randomized in each cohort. The first 12 subjects will be randomized to the KPL-716 and the placebo arms in a 3:1 randomization ratio. The rest of the 14 subjects will be randomized in a 1:1 randomization ratio. There will be approximately 16 subjects and 10 subjects randomized to the KPL-716 arm and the placebo arm respectively. Based on a two-sample t-test for the primary efficacy endpoint, change from baseline in weekly average of WI-NRS at Week 8, assumed mean changes of 4 for the KPL-716 arm and
Determination of Sample Size. To reliably detect (p<0.05, statistical power >80%) a more than two-fold increase in the odds of achieving smoking reduction at 24 weeks among the active versus placebo groups, and assuming a smoking reduction rate of 15% in the placebo group versus 25% in the active snus group (corresponding to an odds ratio of 1.9), the target sample size can be estimated at 250 per group, that is, a total size of 500 study participants. In previous randomized studies of NRT versus placebo, NRT has resulted in an increased proportion of participants achieving short- to medium term smoking cessation with odds ratios in the order of 1.5-3. Epidemiological, cross-sectional studies from Sweden have indicated that snus might be more effective as an aid in smoking cessation than NRT. Against this background the mentioned target sample size should provide reasonable statistical power for current purposes.
