Showing papers in "Journal of Biopharmaceutical Statistics in 2012"

Adaptive randomization for clinical trials.

Abstract: In February 2010, the U.S. Food and Drug Administration (FDA, 2010) drafted guidance that discusses the statistical, clinical, and regulatory aspects of various adaptive designs for clinical trials. An important class of adaptive designs is adaptive randomization, which is considered very briefly in subsection VI.B of the guidance. The objective of this paper is to review several important new classes of adaptive randomization procedures and convey information on the recent developments in the literature on this topic. Much of this literature has been focused on the development of methodology to address past criticisms and concerns that have hindered the broader use of adaptive randomization. We conclude that adaptive randomization is a very broad area of experimental design that has important application in modern clinical trials.

Adaptive randomization to improve utility-based dose-finding with bivariate ordinal outcomes.

Abstract: A sequentially outcome-adaptive Bayesian design is proposed for choosing the dose of an experimental therapy based on elicited utilities of a bivariate ordinal (toxicity, efficacy) outcome. Subject to posterior acceptability criteria to control the risk of severe toxicity and exclude unpromising doses, patients are randomized adaptively among the doses having posterior mean utilities near the maximum. The utility increment used to define near-optimality is nonincreasing with sample size. The adaptive randomization uses each dose's posterior probability of a set of good outcomes, defined by a lower utility cutoff. Saturated parametric models are assumed for the marginal dose-toxicity and dose-efficacy distributions, allowing the possible requirement of monotonicity in dose, and a copula is used to obtain a joint distribution. Prior means are computed by simulation using elicited outcome probabilities, and prior variances are calibrated to control prior effective sample size and obtain a design with good operating characteristics. The method is illustrated by a Phase I/II trial of radiation therapy for children with brainstem gliomas.

Simultaneous generation of binary and normal data with specified marginal and association structures.

Abstract: Situations in which multiple outcomes and predictors of different distributional types are collected are becoming increasingly common in biopharmaceutical practice, and joint modeling of mixed types has been gaining popularity in recent years. Evaluation of various statistical techniques that have been developed for mixed data in simulated environments necessarily requires joint generation of multiple variables. This article is concerned with building a unified framework for simulating multiple binary and normal variables simultaneously given marginal characteristics and association structure via combining well-established results from the random number generation literature. We illustrate the proposed approach in two simulation settings where we use artificial data as well as real depression score data from psychiatric research, demonstrating a very close resemblance between the specified and empirically computed statistical quantities of interest through descriptive and model-based tools.

On the independence of data monitoring committee in adaptive design clinical trials.

Abstract: In clinical trials, an independent data monitoring committee (DMC) is often established to perform both ongoing safety data monitoring and interim efficacy analysis. These evaluations are performed in a blinded fashion in order to avoid possible operational biases that may be introduced to the trial after the review of the data. The DMCs for clinical trials using adaptive design methods are also positioned to implement the adaptation decision according to the prospective adaptation algorithm. While the DMC plays an important role in maintaining the validity and integrity of the intended clinical trial, adaptive design clinical trials trigger a greater role and increased responsibility for the DMC. To assist the sponsor in establishing a DMC, the U.S. Food and Drug Administration (FDA) published a draft guidance entitled Establishment and Operation of Clinical Trial Data Monitoring Committees in 2006. In this article, the composition, role/responsibility, and function/activity of a DMC are described. Concerns of the additional responsibilities of the DMC for adaptive design clinical trials are addressed. Although the intention of the DMC is well-intentioned, controversial issues inevitably occur. These controversial issues include, but are not limited to, (1) the challenge of the independence of a DMC and (2) the issue regarding the direct communication between the DMC and the FDA. Discussion of controversial issues and practical issues are also provided.

Correcting Overall Survival for the Impact of Crossover Via a Rank-Preserving Structural Failure Time (RPSFT) Model in the RECORD-1 Trial of Everolimus in Metastatic Renal-Cell Carcinoma

Abstract: Clinical trials in oncology often allow patients randomized to placebo to cross over to the active treatment arm after disease progression, leading to underestimation of the treatment effect on overall survival as per the intention-to-treat analysis We illustrate the statistical aspects and practical use of the rank-preserving structural failure time (RPSFT) model with the Fleming-Harrington family of tests to estimate the crossover-corrected treatment effect, and to assess its sensitivity to various weighting schemes in the RECORD-1 trial The results suggest that the benefit demonstrated in progression-free survival is likely to translate into a robust overall survival benefit

Average Cost-Effectiveness Ratio with Censored Data

Abstract: In cost-effectiveness analysis, interest could lie foremost in the incremental cost-effectiveness ratio (ICER), which is the ratio of the incremental cost to the incremental benefit of two competing interventions. The average cost-effectiveness ratio (ACER) is the ratio of the cost to benefit of an intervention without reference to a comparator. A vast literature is available for statistical inference of the ICERs, but limited methods have been developed for the ACERs, particularly in the presence of censoring. Censoring is a common feature in prospective studies, and valid analyses should properly adjust for censoring in cost as well as in effectiveness. In this article, we propose statistical methods for constructing a confidence interval for the ACER from censored data. Different methods-Fieller, Taylor, bootstrap-are proposed, and through simulation studies and data analysis, we address the performance characteristics of these methods.

Minimizing the Maximum Expected Sample Size in Two-Stage Phase II Clinical Trials with Continuous Outcomes

Abstract: Two-stage designs are commonly used for Phase II trials. Optimal two-stage designs have the lowest expected sample size for a specific treatment effect, for example, the null value, but can perform poorly if the true treatment effect differs. Here we introduce a design for continuous treatment responses that minimizes the maximum expected sample size across all possible treatment effects. The proposed design performs well for a wider range of treatment effects and so is useful for Phase II trials. We compare the design to a previously used optimal design and show it has superior expected sample size properties.

Drug Dosage Individualization Based on a Random-Effects Linear Model

Abstract: This article investigates drug dosage individualization when the patient population can be described with a random-effects linear model of a continuous pharmacokinetic or pharmacodynamic response. Specifically, we show through both decision-theoretic arguments and simulations that a published clinical algorithm may produce better individualized dosages than some traditional methods of therapeutic drug monitoring. Since empirical evidence suggests that the linear model may adequately describe drugs and patient populations, and linear models are easier to handle than the nonlinear models traditionally used in population pharmacokinetics, our results highlight the potential applicability of linear mixed models to dosage computations and personalized medicine.

Design and Sample Size Considerations for Simultaneous Global Drug Development Program

Abstract: Due to the potential impact of ethnic factors on clinical outcomes, the global registration of a new treatment is challenging. China and Japan often require local trials in addition to a multiregional clinical trial (MRCT) to support the efficacy and safety claim of the treatment. The impact of ethnic factors on the treatment effect has been intensively investigated and discussed from different perspectives. However, most current methods are focusing on the assessment of the consistency or similarity of the treatment effect between different ethnic groups in exploratory nature. In this article, we propose a new method for the design and sample size consideration for a simultaneous global drug development program (SGDDP) using weighted z-tests. In the proposed method, to test the efficacy of a new treatment for the targeted ethnic (TE) group, a weighted test that combines the information collected from both the TE group and the nontargeted ethnic (NTE) group is used. The influence of ethnic factors and local medical practice on the treatment effect is accounted for by down-weighting the information collected from NTE group in the combined test statistic. This design controls rigorously the overall false positive rate for the program at a given level. The sample sizes needed for the TE group in an SGDDP for three most commonly used efficacy endpoints, continuous, binary, and time-to-event, are then calculated.

Regulatory Considerations in the Design of Comparative Observational Studies Using Propensity Scores

Abstract: In the evaluation of medical products, including drugs, biological products, and medical devices, comparative observational studies could play an important role when properly conducted randomized, well-controlled clinical trials are infeasible due to ethical or practical reasons. However, various biases could be introduced at every stage and into every aspect of the observational study, and consequently the interpretation of the resulting statistical inference would be of concern. While there do exist statistical techniques for addressing some of the challenging issues, often based on propensity score methodology, these statistical tools probably have not been as widely employed in prospectively designing observational studies as they should be. There are also times when they are implemented in an unscientific manner, such as performing propensity score model selection for a dataset involving outcome data in the same dataset, so that the integrity of observational study design and the interpretability of outcome analysis results could be compromised. In this paper, regulatory considerations on prospective study design using propensity scores are shared and illustrated with hypothetical examples.

Power and sample size calculation for microarray studies.

Abstract: Microarray is a technology to screen a large number of genes to discover those differentially expressed between clinical subtypes or different conditions of human diseases. Gene discovery using microarray data requires adjustment for the large-scale multiplicity of candidate genes. The family-wise error rate (FWER) has been widely chosen as a global type I error rate adjusting for the multiplicity. Typically in microarray data, the expression levels of different genes are correlated because of coexpressing genes and the common experimental conditions shared by the genes on each array. To accurately control the FWER, the statistical testing procedure should appropriately reflect the dependency among the genes. Permutation methods have been used for accurate control of the FWER in analyzing microarray data. It is important to calculate the required sample size at the design stage of a new (confirmatory) microarray study. Because of the high dimensionality and complexity of the correlation structure in micro...

New Nonparametric Confidence Intervals for the Youden Index

Abstract: The Youden index, a main summary index for the receiver operating characteristic (ROC) curve, is a comprehensive measurement for the effectiveness of a diagnostic test. For a continuous-scale diagnostic test, the optimal cut point for positive disease is the cut point leading to the maximization of the sum of sensitivity and specificity. Finding the Youden index of the test is equivalent to maximize the sum of sensitivity and specificity for all the possible values of the cut point. In this paper, we propose new nonparametric confidence intervals for the Youden index. Extensive simulation studies are conducted to compare the relative performance of the new intervals with the existing parametric interval for the index. Our simulation results indicate that the newly developed nonparametric intervals are competitive with the existing parametric interval when the underlying distributions are correctly specified, and they outperform the existing parametric interval when the underlying distributions are misspecified. The new intervals are robust and easy to implement in practice. A real example is also used to illustrate the application of the proposed intervals.

A Bayesian Inference of P(π1 > π2) for Two Proportions

Abstract: The statistical inference concerning the difference between two independent binominal proportions is often discussed in medical and statistical literature. However, such discussions are often based on the frequentist viewpoint rather than the Bayesian viewpoint. In this article, we propose an index θ =P(π1, post > π2, post ), where π1, post and π2, post denote binominal proportions following posterior density. We provide approximate and exact expressions for θ by using the beta prior. We also present the results of actual clinical trials to show the utility of θ. Our findings suggest that θ can potentially provide useful information in a clinical trial.

Assessment of Regional Treatment Effect in a Multiregional Clinical Trial

Abstract: The 11th question-and-answer document (Q&A) for ICH E5 (1998) was published in 2006. This Q&A describes points to consider for evaluating the possibility of bridging among regions by a multiregional trial. The primary objective of a multiregional bridging trial is to show the overall efficacy of a drug in all participating regions while also evaluating the possibility of applying the overall trial results to each region. To apply the overall results to a specific region, it suggested that the results in that region should be consistent with the overall results. The Japanese Ministry of Health, Labor, and Welfare (MHLW) published the "Basic Principles on Global Clinical Trials" guidance document (2007) and proposed two methods to support the bridging claims. Due to the limited sample sizes allocated to the region, the regular interaction test for treatment by region is not practical. On the other hand, the sample size requirement for the Japanese region as described in Uyama et al. (2005) and Uesaka (2009) is to satisfy an 80% or greater power for the Japanese region, conditioning on the effect of the overall global trial. Quan et al. (2010) further extended the results to trials with various endpoints. Ko, Tsou, Liu and Hsiao (2010) focused on a specific region and established statistical criteria for consistency between the region of interest and overall results. The proposed method was based on the assumption that true effect size is uniform across regions. In this article, we propose to analyze a completed multiregional trial for any specific regional effect by controlling the type I error rate adjusted for the regional sample size and the planned power of the global trial. Accordingly, in order to attain the approval for a specific region, we propose to determine the sample size requirement for the specific regions using the overall power planned and a regional acceptable type I error rate.

Doubly Randomized Delayed-Start Design for Enrichment Studies with Responders or Nonresponders

Abstract: High placebo response has been a major source of bias and is difficult to deal with in many central nervous system (CNS) clinical trials. This bias has led to a high failure rate in mood disorder trials even with known effective drugs. For cancer trials, the traditional parallel group design biases the inference on the maintenance effect of the new drug with the traditional time-to-treatment failure analysis. To minimize bias, we propose a doubly randomized delayed-start design for clinical trials with enrichment. The design consists of two periods. In the first period, patients can be randomized to receive several doses of a new drug or a control. In the second period, control patients of the first period of an enriched population can be rerandomized to receive the same or fewer doses of the new drug or to continue on the control. Depending on the clinical needs, different randomization ratios can be applied to the two periods. The essential feature is that the design is naturally adaptive because of the randomization for the second period. As a result, other aspects of the second period, such as the sample size, can be modified adaptively when an interim analysis is set up for the first period. At the end of the trial, response data from both randomizations are combined in an integrated analysis. Because of the enrichment in the second period, the design increases the probability of trial success and, in addition, reduces the required sample size. Thus, for clinical development, the design offers greater efficiency.

Multiregional clinical trials: Japanese perspective on drug development strategy and sample size for Japanese subjects.

Abstract: Multiregional clinical trials including Japanese subjects are playing a key role in new drug development in Japan. In addition to the consideration of differences in intrinsic and extrinsic ethnic factors, deciding the sample size of Japanese subjects is an important issue when a multiregional clinical trial is intended to be used for Japanese submission. Accumulated experience suggests that there are several points to consider, such as the basic principles described in the guidance document, drug development strategy, trial phase, and disease background. The difficulty of interpreting the results of Japanese trials should also be considered.

Flexible Analytical Methods for Adding a Treatment Arm Mid-Study to an Ongoing Clinical Trial

Abstract: It is not uncommon to have experimental drugs under different stages of development for a given disease area Methods are proposed for use when another treatment arm is to be added mid-study to an ongoing clinical trial Monte Carlo simulation was used to compare potential analytical approaches for pairwise comparisons through a difference in means in independent normal populations including (1) a linear model adjusting for the design change (stage effect), (2) pooling data across the stages, or (3) the use of an adaptive combination test In the presence of intra-stage correlation (or a non-ignorable fixed stage effect), simply pooling the data will result in a loss of power and will inflate the type I error rate The linear model approach is more powerful, but the adaptive methods allow for flexibility (re-estimating sample size) The flexibility to add a treatment arm to an ongoing trial may result in cost savings as treatments that become ready for testing can be added to ongoing studies

Evaluation of the pre-posterior distribution of optimized sampling times for the design of pharmacokinetic studies

Abstract: Information theoretic methods are often used to design studies that aim to learn about pharmacokinetic and linked pharmacokinetic–pharmacodynamic systems. These design techniques, such as D-optimality, provide the optimum experimental conditions. The performance of the optimum design will depend on the ability of the investigator to comply with the proposed study conditions. However, in clinical settings it is not possible to comply exactly with the optimum design and hence some degree of unplanned suboptimality occurs due to error in the execution of the study. In addition, due to the nonlinear relationship of the parameters of these models to the data, the designs are also locally dependent on an arbitrary choice of a nominal set of parameter values. A design that is robust to both study conditions and uncertainty in the nominal set of parameter values is likely to be of use clinically. We propose an adaptive design strategy to account for both execution error and uncertainty in the parameter values. In...

Design and Evaluation of Multiregional Trials with Heterogeneous Treatment Effect Across Regions

Abstract: To speed up drug development to allow faster access to medicines for patients globally, conducting multiregional trials incorporating subjects from many countries around the world under the same protocol may be desired. Several statistical methods have been proposed for the design and evaluation of multiregional trials. However, in most of the recent approaches for sample size determination in multiregional trials, a common treatment effect of the primary endpoint across regions is usually assumed. In practice, it might be expected that there is a difference in treatment effect due to regional difference (e.g., ethnic difference). In this article, a random effect model for heterogeneous treatment effect across regions is proposed for the design and evaluation of multiregional trials. We also address consideration of the determination of the number of subjects in a specific region to establish the consistency of treatment effects between the specific region and the entire group.

Decision rules and associated sample size planning for regional approval utilizing multiregional clinical trials.

Abstract: Multiregional clinical trials provide the potential to make safe and effective medical products simultaneously available to patients globally. As regulatory decisions are always made in a local context, this poses huge regulatory challenges. In this article we propose two conditional decision rules that can be used for medical product approval by local regulatory agencies based on the results of a multiregional clinical trial. We also illustrate sample size planning for such trials.

A Note on Power and Sample Size Calculations for the Kruskal–Wallis Test for Ordered Categorical Data

Abstract: Although the Kruskal-Wallis test has been widely used to analyze ordered categorical data, power and sample size methods for this test have been investigated to a much lesser extent when the underlying multinomial distributions are unknown This article generalizes the power and sample size procedures proposed by Fan et al ( 2011 ) for continuous data to ordered categorical data, when estimates from a pilot study are used in the place of knowledge of the true underlying distribution Simulations show that the proposed power and sample size formulas perform well A myelin oligodendrocyte glycoprotein (MOG) induced experimental autoimmunce encephalomyelitis (EAE) mouse study is used to demonstrate the application of the methods

A Review of: “ An R Companion to Applied Regression , Second Edition, by J. Fox and S. Weisberg”

Abstract: There seems to be a new book on R published every day, and this is one of the better ones. Nearly all such books are reasonably advanced, with Verzani (2005) the notable exception. The reader of th...

On Sample Size Calculation for Comparing Survival Curves Under General Hypothesis Testing

Abstract: The log-rank test is commonly used to test the equivalence of two survival distributions under right censoring. Jung et al. (2005) proposed a modified log-rank test for noninferiority trials and its corresponding sample size calculation. In this article, we extend the use of the modified log-rank test for clinical trials with various types of nonconventional study objectives and propose its sample size calculation under general null and alternative hypotheses. The proposed formula is so flexible that we can specify any survival distributions and accrual pattern. The proposed methods are illustrated with designing real clinical trials. Through simulations, the modified log-rank test and the derived formula for sample size calculation are shown to have satisfactory small sample performance.

Designing studies to find early signals of efficacy.

Abstract: We introduce the idea of a design to detect signals of efficacy in early phase clinical trials. Such a design features three possible decisions: to kill the compound; to continue with staged development; or to continue with accelerated development of the compound. We describe how such studies improve the trade-off between the two errors of killing a compound with good efficacy and committing to a complete full development program for a compound that has no efficacy and describe how they can be designed. We argue that such studies could be used to screen compounds at the proof-of-concept state, reduce late Phase 2 attrition, and speed up the development of highly efficacious drugs.

Test Procedures for Disease Prevalence with Partially Validated Data

Abstract: Investigating the prevalence of a disease is an important topic in medical studies. Such investigations are usually based on the classification results of a group of subjects according to whether they have the disease. To classify subjects, screening tests that are inexpensive and nonintrusive to the test subjects are frequently used to produce results in a timely manner. However, such screening tests may suffer from high levels of misclassification. Although it is often possible to design a gold-standard test or device that is not subject to misclassification, such devices are usually costly and time-consuming, and in some cases intrusive to the test subjects. As a compromise between these two approaches, it is possible to use data that are obtained by the method of double-sampling. In this article, we derive and investigate four test statistics for testing a hypothesis on disease prevalence with double-sampling data. The test statistics are implemented through both the asymptotic method suitable for lar...

Assessing Interactions for Fixed-Dose Drug Combinations in Tumor Xenograft Studies

Abstract: Statistical methods for assessing the joint action of compounds administered in combination have been established for many years. However, there is little literature available on assessing the joint action of fixed-dose drug combinations in tumor xenograft experiments. Here an interaction index for fixed-dose two-drug combinations is proposed. Furthermore, a regression analysis is also discussed. Actual tumor xenograft data were analyzed to illustrate the proposed methods.

Phase II Cancer Clinical Trials with Heterogeneous Patient Populations

Abstract: The patient population for a Phase II trial often consists of multiple subgroups in terms of risk level. In this case, a popular design approach is to specify the response rate and the prevalence of each subgroup, to calculate the response rate of the whole population by the weighted average of the response rates across subgroups, and to choose a standard Phase II design such as Simon's optimal or minimax design to test the response rate for the whole population. In this case, although the prevalence of each subgroup is accurately specified, the observed prevalence among the accrued patients to the study may be quite different from the expected one because of the small sample size, which is typical in most Phase II trials. The fixed rejection value for a chosen standard Phase II design may be either too conservative (i.e., increasing the false rejection probability of the experimental therapy) if the trial accrues more high-risk patients than expected, or too anti-conservative (i.e., increasing the false acceptance probability of the experimental therapy) if the trial accrues more low-risk patients than expected. We can avoid such problems by adjusting the rejection values, depending on the observed prevalence from the trial. In this paper, we investigate the performance of the flexible designs compared with the standard design with fixed rejection values under various settings.

Extended key-factor/key-stage analysis for longitudinal data.

Abstract: Key-factor/key-stage analysis was originally a descriptive approach to analyze life tables. However, this method can be extended to analyze longitudinal data in pharmaceutical experiments. By dividing the variance into components, the extended key-factor/key-stage analysis indicates which factor is influential, and through which stage the factor generates its influence in determining the outcome of treatments. Such knowledge helps us in constructing a class of nonlinear longitudinal models that can be interpretable than linear models. Example SAS programs and R programs are provided for the calculation. Supplemental materials are available for this article. Go to the publisher's online edition of Journal of Biopharmaceutical Statistics to view the supplemental files.

Statistical Methods for Bridging Studies

Abstract: In 1998, the International Conference on Harmonization (ICH) published a guidance to facilitate the registration of medicines among ICH regions including the European Union, the United States, and Japan by recommending a framework for evaluating the impact of ethnic factors on a medicine's effect, such as its efficacy and safety at a particular dosage and dose regimen (ICH E5, 1998). The purpose of ICH E5 is not only to evaluate the ethnic factor influence on safety, efficacy, dosage, and dose regimen, but also more importantly to minimize duplication of clinical data and allow extrapolation of foreign clinical data to a new region. In this article, statistical methods for evaluation of bridging studies based on the concepts of consistency (Shih, 2001), reproducibility/generalizability (Shao and Chow, 2002), the weighted Z-tests for the design of bridging studies (Lan et al., 2005), and similarity between the new and original region based in terms of positive treatment effect (Hsiao et al., 2007) are studied. The relative merits and disadvantages of these methods are compared by several examples.

A consistency-adjusted strategy for accommodating an underpowered primary endpoint.

Abstract: A clinical trial might involve more than one clinically important endpoint, each of which can characterize the treatment effect of the experimental drug under investigation. Underlying the concept of using such endpoints interchangeably to establish an efficacy claim, or pooling different endpoints to constitute a composite endpoint, is the assumption that findings from such endpoints are consistent with each other. While such an assumption about consistency of efficacy findings appears to be intuitive, it is seldom considered in the design and analysis literature of clinical trials with multiple endpoints. Failure to account for consistency of efficacy findings of two candidate endpoints to establish efficacy, at the design stage, has led to difficulties in interpreting study findings. This article presents a flexible testing strategy for accommodating findings of an alternative to the designated primary endpoint (or a subgroup) to support an efficacy claim. The method is built on the following two premi...