ADJUSTING PAIRWISE NONPARAMETRIC EQUIVALENCE HYPOTHESIS TESTS AND CONFIDENCE INTERVALS FOR PERIOD EFFECTS IN 3×3 CROSSOVER TRIALS

In pharmacokinetic and pharmacodynamic 3×3 crossover trials, average bioequivalence and noninferiority between treatments need to be only assessed pairwise in most cases. Due to the restricted number of subjects in such trials, normal distribution assumptions cannot be checked and frequently outliers are encountered, so that a nonparametric approach is more adequate. Therefore, to assess average bioequivalence or noninferiority, a new method is proposed to derive period adjusted nonparametric confidence intervals for pairwise treatment differences.     

ON SAMPLE SIZE CALCULATION BASED ON ODDS RATIO IN CLINICAL TRIALS

Sample size calculation formulas for testing equality, noninferiority, superiority, and equivalence based on odds ratio were derived under both parallel and one-arm crossover designs. An example concerning the study of odds ratio between a test compound (treatment) and a standard therapy (control) for prevention of relapse in subjects with schizophrenia and schizoaffective disorder is presented to illustrate the derived formulas for sample size calculation for various hypotheses under both a parallel design and a crossover design. Simulations were performed to assess the adequacy of the sample size calculation formulas. Simulation results were given at the end of the paper.     

The Role of Multiple Imputation in Noninferiority Trials for Binary Outcomes

We consider the role of multiple imputation (MI) when analyzing noninferiority (NI) clinical trials with missing data. When the endpoint is measured longitudinally, direct-likelihood methods can be used. In this article, the focus is on the situation in which the endpoint is not measured longitudinally but other relevant data are measured at or after baseline prior to planned collection of the primary endpoint data. Simulation results are presented for various scenarios based on the missingness mechanism, the dropout rate, and the size of NI margin. When the endpoint is binary, the ratio of the amount of missing data to the noninferiority margin will affect the operating characteristics of any analysis strategy (whether imputation based or not), an issue that is unique to noninferiority trials. Biased estimates of treatment effect under missingness, not completely at random, may arise when using a misspecified imputation model lacking treatment effect, resulting in substantially inflated Type I error rates in noninferiority trials by making the two groups appear more similar, opposite the usual impact in superiority trials. As in superiority trials, MI will have most benefit when data are missing at random, and the important predictor variables are included in the imputation model.     

Gabapentin versus lamotrigine monotherapy: a double-blind comparison in newly diagnosed epilepsy

PURPOSE: This randomised, double-blind study compared the newer antiepileptic drugs (AEDs) gabapentin (GBP) and lamotrigine (LTG) as monotherapy in newly diagnosed epilepsy. METHODS: Patients with partial seizures with and/or without secondary generalization or primary generalized tonic-clonic seizures were randomized to either GBP or LTG. During 2- and 6-week titration periods, respectively, GBP dosage reached 1,800 mg/day, and LTG, 150 mg/day. In the subsequent 24-week maintenance phase, the dose could be adjusted based on seizure control or adverse events between 1,200 and 3,600 mg/day for GBP and 100 and 300 mg/day for LTG. The primary end point was time to exit, a composite of efficacy and tolerability. Evaluable patients were used for the primary efficacy analysis, whereas tolerability was examined on an intent-to-treat basis. RESULTS: A total of 309 patients was randomized, and 291 (148 GBP, 143 LTG) were included in the evaluable population. Nineteen patients in each group had an exit event. The median time to exit was 69 days for GBP and 48 days for LTG. The hazard ratio was estimated as 1.043 (90% confidence intervals, 0.602-1.809). Overall, 106 (71.6% of the evaluable population) GBP-treated and 96 (67.1%) LTG-treated patients completed the study. Of those, 80 (75.5%) patients taking GBP and 73 (76.0%) taking LTG remained seizure free during the final 12 weeks of treatment. Only 14 (8.9%) GBP-treated patients and 15 (9.9%) LTG-treated patients withdrew because of study drug-related adverse events. CONCLUSIONS: GBP and LTG monotherapy were similarly effective and well tolerated in patients with newly diagnosed epilepsy.     

Likelihood ratio and a Bayesian approach were superior to standard noninferiority analysis when the noninferiority margin varied with the control event rate

OBJECTIVE: To present and compare three statistical approaches for analyzing a noninferiority trial when the noninferiority margin depends on the control event rate. STUDY DESIGN AND SETTING: In noninferiority trials with a binary outcome, the noninferiority margin is often defined as a fixed delta, the largest clinically acceptable difference in event rates between treatment groups. An alternative and more flexible approach is to allow delta to vary according to the true event rate in the control group. The appropriate statistical method for evaluating noninferiority with a variable noninferiority margin is not apparent. Three statistical approaches are proposed and compared: an observed event rate (OER) approach based on equating the true control rate to the observed rate, a Bayesian approach, and a likelihood ratio test. RESULTS AND CONCLUSIONS: Simulations studies indicate that the proportion of trials in which noninferiority was erroneously demonstrated was higher for the OER approach than with the Bayesian and likelihood ratio approaches. In some cases, the Type I error rate exceeded 10% for the OER approach. The OER approach is not recommended for the analysis of noninferiority trials with a variable margin of equivalence. The Bayesian and likelihood ratio methods yielded better operating characteristics.     

Issues in the design, analysis and interpretation of condom functionality studies

Background

Male condom functionality studies are typically crossover trials in which enrolled couples use both experimental and latex control condoms for sexual intercourse. Noninferiority of the experimental type is assessed using confidence intervals for differences in breakage and slippage probabilities. Seemingly straightforward, the design, analysis and interpretation of functionality studies are complicated by the choice of noninferiority criterion, study population and the potential for learning effects.

Methods

Power calculations, secondary data analyses and simulations were used to illustrate concerns and make recommendations.

Results

The probability of failure can be too low to draw meaningful conclusions in certain population subgroups. Learning effects among inexperienced users can exaggerate differences in performance and undermine power. A product which is, on average, inferior to latex may still be a viable prophylactic for a large percentage of couples.

Conclusions

Heterogeneity of failure probabilities, combined with small acceptable differences in performance, requires care when selecting study participants. Pilot data, adequate training on condom use and reasonable expectations regarding performance of a new condom type are essential to maximizing the chance of identifying a noninferior product.     

Demonstrating Biosimilarity via Equivalence in Clinical Trials

The opportunities for biosimilar medicines have stimulated much legal and regulatory debates and actions around the world, most notably the passing of the Biologics Price Competition and Innovation Act in the United States in 2009. A key difference between the development of a biosimilar product versus a generic chemical entity is the requirement for well-controlled clinical studies to demonstrate similarity in efficacy and safety. The main objective of this article is to extend the clinical study design methods commonly used in noninferiority trials to equivalence trials within the context of biosimilar product development. We extend the synthesis method to the equivalence setting and provide sample size considerations. We show that while an equivalence trial in general requires a larger sample size than a noninferiority trial, the difference may not be substantial depending on the significance level required for the equivalence trial.     

Exact and Approximate Power and Sample Size Calculations for Analysis of Covariance in Randomized Clinical Trials With or Without Stratification

ABSTRACT

Analysis of covariance (ANCOVA) is commonly used in the analysis of randomized clinical trials to adjust for baseline covariates and improve the precision of the treatment effect estimate. We derive the exact power formulas for testing a homogeneous treatment effect in superiority, noninferiority, and equivalence trials under both unstratified and stratified randomizations, and for testing the overall treatment effect and treatment × stratum interaction in the presence of heterogeneous treatment effects when the covariates excluding the intercept, treatment, and prestratification factors are normally distributed. These formulas also work very well for nonnormal covariates. The sample size methods based on the normal approximation or the asymptotic variance generally underestimate the required size. We adapt the recently developed noniterative and two-step sample size procedures to the above tests. Both methods take into account the nonnormality of the t statistic, and the lower order variance term commonly ignored in the sample size estimation. Numerical examples demonstrate the excellent performance of the proposed methods particularly in small samples. We revisit the topic on the prestratification versus post-stratification by comparing their relative efficiency and power. Supplementary materials for this article are available online.     

PISC Expert Team White Paper: Toward a Consistent Standard of Evidence When Evaluating the Efficacy of an Experimental Treatment From a Randomized,Active-Controlled Trial

The double-blind placebo-controlled trial is the established standard for determining the efficacy of an experimental treatment. However, there are circumstances where the use of a placebo is unethical or impractical, and active-controlled trials are a common alternative. In an active-controlled trial, the objective is typically to show that the effect of the experimental treatment is within some prespecified margin of the control effect. The margin is often chosen specifically to guarantee 50% or 75% preservation of the control effect over placebo. An implicit assumption is that a higher standard of efficacy is required when a new treatment is evaluated in an active-controlled trial. In this article, we argue that standards based on margins and/or percent preservation are inherently arbitrary and lacking in objective clinical or scientific justification. The use of these ad hoc standards introduces logical inconsistencies for regulatory evaluation such that safe and effective treatments may be denied regulatory approval. We therefore argue that active-controlled trials should not require an arbitrarily higher standard of evidence than placebo-controlled trials. We discuss how “synthesis” analyses combining the results of both an active-controlled trial and one or more historical trials can address methodological challenges and provide for adequate estimation and testing of an experimental treatment effect relative to placebo.     

Genomic Classifier for Patient Enrichment: Misclassification and Type I Error Issues in Pharmacogenomics Noninferiority Trial

Unlike the phenotypic characteristics, a genomic diagnostic assay is needed to assess whether a patient truly has the genomic characteristics of interest for therapeutic investigation. However, in reality, the diagnostic assay is not perfect without misclassification error. We consider a targeted or enrichment pharmacogenomics clinical trial with a noninferiority objective. We assume the probability of response of the misclassified subjects is identical in the treated and control groups because the blinding should yield no preferential misclassification. We show that the maximum Type I error rate associated with falsely concluding noninferiority can be substantially liberal if the accuracy of the diagnostic assay is much less than perfect. When the diagnostic assay has moderate to high sensitivity and specificity, a reasonable range of false nondisease prediction rate among test positives can be identified where the maximum Type I error rate inflation is much less. To achieve a prescribed level of positive predictive value, the feasibility of a genomic biomarker for enrichment in a noninferiority study is a function of disease prevalence. Given the evidential standard of much less inflation of Type I error rate can be shown, a genomic biomarker, if qualified, may be useful as an adjunct tool for patient selection.