Modelling and design of cross-over trials

There are many diseases and conditions that can be studied using a cross-over clinical trial, where the subjects receive sequences of treatments. The treatments are then compared using the repeated measurements taken 'within' subjects. The actual plan or design of the trial is usually obtained by consulting a published table of designs or by applying relatively simple rules such as using all possible permutations of the treatments. However, there is a danger is this approach because the model assumed for the data when the tables or rules were constructed may not be appropriate for the new trial being planned. Also, there may be restrictions in the new trial on the number of treatment sequences that can be used or on the number of periods of treatment particular subjects can be given. Such restrictions may mean that a published design of the ideal size cannot be found unless compromises are made. A better approach is to make the design satisfy the objectives of the trial rather than vice versa. In this paper we describe an approach to constructing such tailor-made designs which we hope will lead to ill-fitting 'off the peg' designs being a thing of the past. We use a computer algorithm to search for optimal designs and illustrate it using a number of examples. The criterion of optimality used in this paper is A-optimality but our approach is not restricted to one particular criterion. The model used in the search for the optimal design is chosen to suit the nature of the trial at hand and as an example a variety of models for three treatments are considered. We also illustrate the construction of designs for the comparison of two active treatments and a placebo where it can be assumed that the carry-over effects of the active treatments are similar. Finally, we illustrate an augmentation of a design that could arise when the objectives of a trial change.     

An evaluation of phase I cancer clinical trial designs

Phase I clinical trials are designed to identify an appropriate dose for experimentation in phase II and III studies. I present the results from a simulation study to evaluate the performance of nine phase I designs involving the standard design, the two-stage modified Storer's design, the two-stage Korn's design, the one-stage modified continual reassessment method (CRM) designs, and the two-stage modified CRM designs. I compare the performance of the above phase I designs in terms of the following criteria: (i) the proportion of the recommended maximum tolerated dose (MTD) at each dose level; (ii) the proportion of patients treated at each dose level; (iii) the average number of patients to complete the trial; (iv) the probability of toxicity observed; and (v) the average number of cohorts to complete the trial. In general, the one-stage modified CRM II and CRM III designs perform well compared with the other designs considered in this study. The one-stage modified CRM II and III designs require much fewer numbers of cohorts than do the two-stage modified CRM II and III designs. The one-stage modified CRM II and III designs avoid the criticisms of the original CRM by reducing the average number of cohorts and toxicity incidences, while estimating the MTD more accurately than does the standard design.     

Statistical approaches to trial durations in episodic affective illness

In light of the high variability in illness characteristics and patterns among patients with bipolar illness, parallel group designs present severe methodologic difficulties. Crossover, off-on-off-on (B-A-B-A), and other individualized designs may be a useful substitute, but no consensus exists about how to estimate the individual trial durations required in these instances. Several methods for determining optimum trial lengths in crossover designs are presented, illustrated, and discussed. These include: chi-square (chi2) for the expected versus observed number of either episodes or days well; exceeding two standard deviations for average duration of episodes or euthymic intervals; or the Sequential Probability Ratio Test (SPRT), which detects when mean values differ from prior statistical expectations. Each method was applied to three demonstration cases using data from actual clinical trials of three patients with different patterns of recurrent affective illness. Each method detected changes in illness severity, although different tests appeared to be sensitive to differing cycle patterns in the patients illustrated. We suggest that these types of analyses and others can be used as indicator statistics to augment global impressions and clinical judgment, and to assist in determining individualized trial durations, both in formal clinical trials and in clinical treatment settings. Once individual responsivity is confirmed with an appropriate interplay of trial design and statistical analysis, the percentage response in a given population can then be compared to other agents or in other populations. Moreover, meta-analytic techniques based on addition of z scores from individuals' effect sizes can then be used to assess overall significance of a drug effect in a given population or subpopulation. The need for further development of appropriate and alternate study designs and analysis methods for bipolar illness is highlighted. Approaches to estimating required trial durations in individuals with different cycle frequencies in crossover and B-A-B-A designs constitute one element of that exploration.     

Placebo washout in trials of antipsychotic drugs

For antipsychotic phase 3 clinical trials, we compare the relative merits of a placebo washout period with an alternate design strategy using a low-dose antipsychotic treatment. Evaluations are made with respect to the achievement of specific clinical trial design goals including the effect on power for detecting between-treatment and within-treatment pre-post differences. The relative merits of these two designs are discussed separately for those patients who enter the initial leadin period after withdrawal from previous antipsychotic medication and for those not on medication immediately before that period.     

Traditional and alternative research designs and methods in clinical pediatric psychopharmacology

OBJECTIVE: To review both traditional and alternative research designs and methods in pediatric psychopharmacology. METHOD: Study designs used in clinical trials in psychiatry were selected for review with the special considerations of pediatric psychopharmacological trials in mind. Where possible, a reference to a specific published trial in pediatric psychopharmacology was provided for each design. RESULTS: It appears that pediatric psychopharmacology trials require a relatively greater flexibility in design to avoid significant study biases. Alternative research designs may be preferable to the traditional approaches, particularly when the use of the latter may raise important issues of feasibility. CONCLUSIONS: Evolution of a clinical research hypothesis to a study protocol in pediatric psychopharmacology is a manageable task if one keeps in mind essential elements such as selection of study design, population, sample size, treatment duration, and efficacy assessments.     

Simulation as a design tool for phase I/II clinical trials: an example from bone marrow transplantation

We discuss the design and analysis of a proposed phase I/II clinical trial in bone marrow transplantation whereby dose modifications that decrease the risk of one complication increase the risk of another. Trials of this type are carried out to determine whether a dose can be found that balances the risks of each complication. Three different scenarios describing potential relationships between each risk and the treatment dose are postulated. The scenarios encompass both favorable situations in which several acceptable doses exist and unfavorable situations in which no acceptable dose exists. The operating characteristics of three sequentially developed trial designs were examined by simulation under each dose-response scenario. The first design was derived from seemingly reasonable rules, but simulations showed that performance fell far short of what was desired, thus motivating modifications. Subsequent designs showed improved performance characteristics compared to the original design. Without close examination of the operating characteristics, the original design would have been implemented, leading to high risk of an erroneous conclusion.     

Conditional power calculations for early acceptance of H0 embedded in sequential tests

For ethical and efficiency concerns one often wishes to design a clinical trial to stop early if there is a strong treatment effect or if there is strong evidence of no treatment effect. There is a large literature to address the design of sequential trials for detecting treatment differences. There has been less attention paid to the design of trials for detecting lack of a treatment difference and most of the designs proposed have been ad hoc modifications of the traditional designs. In the context of fixed sample tests, various authors have proposed basing the decision to stop in favour of the null hypothesis, H0, on conditional power calculations for the end of the trial given the current data. Here I extend this procedure to the popular sequential designs: the O'Brien-Fleming test and the repeated significance test. I derive explicit boundaries for monitoring the test statistic useful for visualizing the impact of the parameters on the operating characteristics of the tests and thus for the design of the tests. Also, they facilitate the use of boundary crossing methods for approximations of power. I derive appropriate boundaries retrospectively for two clinical trials: one that concluded with no treatment difference (AIDS Clinical Trials Group protocol 118) and one that stopped early for positive effect (Beta-Blocker Heart Attack Trial). Finally, I compare the procedures based on the different upper boundaries and assess the impact of allowing for early stopping in favour of H0 in numerical examples.     

Clinical research designs for emerging treatments for Alzheimer disease: moving beyond placebo-controlled trials

The design of clinical trials must evolve as new therapies become available. The demonstrated efficacy and clinical use of donepezil and vitamin E for Alzheimer disease (AD) has shifted the options for AD research design. There is now a compelling case for alternatives to trials that include a treatment arm with no active therapy (ie, a placebo control). With an existing therapy, such as donepezil or vitamin E, new drugs that are clearly superior to those drugs should be sought. Combination therapy is a likely strategy for the future, implying that clinical trials, if possible, should replicate actual practice. The long duration of future AD trials also will make placebo-controlled trials more difficult to justify and more difficult to recruit for. Add-on or active-control designs represent the alternative approaches. We believe that definitive clinical trials of new AD drugs that use one or the other of these designs would be more likely to bring about therapeutic advances than would comparisons with inactive treatments. Our argument is not a general rejection of placebo-control designs. Our recommendations apply only to the circumstances in which the field of AD drug therapy now finds itself.     

Bayesian interim analysis of phase II cancer clinical trials

Many popular sequential phase II clinical trial designs optimize some criterion subject to constraints on the error probabilities at null and alternative values of the response rate. Such designs may forfeit optimality if one fails to conduct analyses strictly according to plan. Moreover, a decision, say, to accept the experimental therapy at one interim analysis does not necessarily imply the same degree of evidence as the same decision when made at another analysis. I propose an alternative design that bases decisions on the ability of the data to persuade either a sceptic or an enthusiast. My standard of evidence, called the persuasion probability, is based on the Bayesian posterior probability that the experimental treatment is superior to the standard. The design calls for termination at any interim analysis at which an observed persuasion probability exceeds its critical value. I investigate the standards of evidence implied by some frequentist procedures and calculate frequentist properties of persuasion-probability designs.     

Optimal flexible designs in phase II clinical trials

In the conduct of a phase II cancer clinical trial, patients usually enter in two stages. If the response rate from the first stage is low, then the study terminates. Within various two-stage designs, Simon proposed the optimal and minimax criteria. In the co-operative group setting, practical considerations make it difficult to arrive at the planned sample size exactly. Green and Dahlberg proposed and compared several flexible designs. In this paper, we explicitly define a flexible design as a collection of two-stage designs where the first stage size is in a set of consecutive values (n1, ..., nk) and the second stage size is also in another set of consecutive values (N1, ..., Nk), and each of k2 possible designs has the same probability of occurrence. We apply Simon's optimal and minimax criteria to flexible designs for phase II trials in order to minimize the number of patients tested on an ineffective drug.     

Advanced regression methods for single-case designs: Studying propranolol in the treatment for agitation associated with traumatic brain injury.

Objective: The use of single-case designs in intervention research is discussed. Regression methods for analyzing data from these designs are considered, and an innovative use of logistic regression to analyze data from a double-blind, randomized clinical trial of propranolol for agitation among persons with traumatic brain injury (TBI) is used. Method: Double-blind, randomized clinical trial performed in an outpatient rehabilitation setting. Participants: Nine men and 4 women with TBI. Results: Logistic models indicated that propranolol was not associated with less agitation for most participants (Φ=-.135; 90% exact confidence interval was -.03     

In situ transmission electron microscopy observations of the formation of self-assembled Ge islands on Si

Advances in our understanding of the pathogenesis of Alzheimer disease (AD) and vascular dementia (VaD) now permit responsible discussion of therapies that may go beyond relief of cognitive and behavioral symptoms and actually slow progression of disease. The mechanisms of neuronal death and the pathologic role of glia are being elucidated, and epidemiologic studies have suggested potential protective value for anti-inflammatory drugs, estrogen, and free-radical scavengers. However, demonstrating disease-modifying drug effects for progressive conditions such as dementia can be a daunting task, fraught with clinical, statistical, and ethical dilemmas. To evaluate trial designs for demonstrating such effects, the International Working Group on Harmonization of Dementia Drug Guidelines (IWG) conducted a symposium at the Sixth International Congress on Alzheimer's Disease and Related Disorders, held July 1998 in Amsterdam. The presentations at the IWG symposium covered the two basic designs currently being used in clinical trials, survival analysis and staggered-start/withdrawal, in addition to clinical data generated from the National Institute on Aging Alzheimer's Disease Cooperative Study vitamin E/selegiline trial in patients with AD and the phase III clinical studies of propentofylline in patients with AD and VaD. It is hoped that this article will open a dialogue among investigators and regulatory authorities regarding appropriate trial designs to support a regulatory claim for disease-modifying effects.     

Enhancing the scientific credibility of single-case intervention research: Randomization to the rescue.

In recent years, single-case designs have increasingly been used to establish an empirical basis for evidence-based interventions and techniques in a variety of disciplines, including psychology and education. Although traditional single-case designs have typically not met the criteria for a randomized controlled trial relative to conventional multiple-participant experimental designs, there are procedures that can be adopted to create a randomized experiment in this class of experimental design. Our two major purposes in writing this article were (a) to review the various types of single-case design that have been and can be used in psychological and educational intervention research and (b) to incorporate randomized experimental schemes into these designs, thereby improving them so that investigators can draw more valid conclusions from their research. For each traditional single-case design type reviewed, we provide illustrations of how various forms of randomization can be introduced into the basic design structure. We conclude by recommending that traditional single-case intervention designs be transformed into more scientifically credible randomized single-case intervention designs whenever the research conditions under consideration permit. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Ex-vivo gene therapy using cytokine-transduced tumor vaccines: molecular and clinical pharmacology

Whether the current generation of cytokine gene-transduced tumor vaccines will show clinical efficacy is under study. Fortunately, the large safety profile so far observed with gene-transduced tumor vaccines can allow outpatient testing in large populations of patients in the adjuvant therapy situation. This will allow large studies statistically powered to see potentially important adjuvant therapy effects in the range that are observed for tamoxifen in breast cancer. For example, the outpatient, adjuvant therapy safety context has been established in the use of GM-CSF gene-transduced autologous prostate cancer vaccines following radical prostatectomy. Similar adjuvant therapy clinical trial efforts are anticipated with allogeneic breast, colon, pancreatic, and ovarian cancer in addition to prostate, renal cell carcinoma, and melanoma. The reverse translation of early clinical data back to basic laboratory research also suggests the field of cytokine gene-transduced tumor vaccine research will remain vibrant. Efforts are currently being directed on optimizing DC activation with polycistronic constructs of cytokine genes, and overexpressing the most relevant tumor-associated peptides. As in the case of antineoplastic drug development, not all lead compounds will become approved drugs in medical oncology. Rigorous yet innovative clinical trial designs will be key to the accelerated identification of cytokine gene-transduced vaccines that improve survival in cancer patients.     

Phase contrasts for multiphase single case intervention designs.

The movement toward evidence-based interventions (EBI) in school psychology places new expectations on scientist-practitioners for evaluating intervention effectiveness (Kratochwill & Stoiber, 2000, 2002). Many such interventions in the school context are conducted for individual students, requiring single case designs. In this paper we provide school psychologists with a technique for assessing intervention effectiveness within complex (multiphase) single case designs, including multiple baseline designs. Phase contrasts can provide the researcher with one or more effect sizes which together summarize the elements of a multiphase design. Phase contrasts can be flexibly specified to match expectations for client performance during particular phases. Additional topics addressed are multiple baseline designs, confidence intervals around effect sizes, and autocorrelation. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

A case for Bayesianism in clinical trials

This paper describes a Bayesian approach to the design and analysis of clinical trials, and compares it with the frequentist approach. Both approaches address learning under uncertainty. But they are different in a variety of ways. The Bayesian approach is more flexible. For example, accumulating data from a clinical trial can be used to update Bayesian measures, independent of the design of the trial. Frequentist measures are tied to the design, and interim analyses must be planned for frequentist measures to have meaning. Its flexibility makes the Bayesian approach ideal for analysing data from clinical trials. In carrying out a Bayesian analysis for inferring treatment effect, information from the clinical trial and other sources can be combined and used explicitly in drawing conclusions. Bayesians and frequentists address making decisions very differently. For example, when choosing or modifying the design of a clinical trial, Bayesians use all available information, including that which comes from the trial itself. The ability to calculate predictive probabilities for future observations is a distinct advantage of the Bayesian approach to designing clinical trials and other decisions. An important difference between Bayesian and frequentist thinking is the role of randomization.     

Effect of sodium nitroprusside on compound action potential thresholds in the gerbil cochlea

OBJECTIVE: Trials that do not allow rejection of the null hypothesis of no treatment effect may have had an inappropriate design. Trials are virtually never assessed for correlation between responses to different treatment modalities. METHODS: Using a hypothetical example and several published studies we examine the influence of correlation levels between treatment modalities on the sensitivity of testing. RESULTS: The level of correlation between responses to different treatment modalities is a major determinant of the sensitivity both of crossover and parallel group clinical trials. CONCLUSIONS: It is very relevant to assess a priori correlation levels between responses to the different treatment modalities of a trial. If a negative correlation is anticipated, a crossover design is likely to lack sensitivity. If a positive correlation is anticipated a parallel-group design seems less appropriate, because it would lack the extra sensitivity of accounting for the positive correlation. Both designs would seem suitable for approximately zero correlations (e.g. comparison vs baseline or vs placebo under the assumption that the number of placebo responders is negligible.     

Crossover designs for clinical trials

I discuss three-period crossover designs for an efficient comparison of two test treatments with special application to clinical trials which often have many practical limitations. In this paper I specify a subset of three-period crossover designs so that the investigators are not left with the problematic two-period two-sequence design, should the trials be terminated after the second period. I show that there is a dramatic reduction in variability for estimating the direct and residual treatment effects in three-period designs compared to two-period designs. I also show that the universally optimal design with ABB and BAA sequences is unsuitable when a complex form of residual effects is suspected, such as the second-order residual effects or treatment by period interactions. The design with ABB, BAA, AAB, and BBA sequences is relatively robust to these uncertain model assumptions. I also discuss missing data problems and conclude that, even with a large proportion of missing values, the three-period design is far more efficient than the two-period design.     

The signals for starvation response are transduced through elevated [Ca2+]i in Dictyostelium cells

Selecting antiretroviral therapies for human immunodeficiency virus type 1-infected persons is complicated by the availability of a vast number of potentially useful drug combinations and by extensive variation among patients in their resistance to various drugs. AIDS clinical trials have used designs in which a handful of drug regimens in a few patient classes can be compared. Here is proposed implementation of innovative designs with factorial structure that permit assessment of many treatment arms and patient classes in a single trial; when and how they can be appropriately used are discussed. These designs are efficient, permit systematic investigation of correlations between genetic mutations and in vivo drug resistance, and provide insight into important drug interactions in people that conventional designs are unable to provide. Through creative application of these designs, identification of superior drug combinations and the science of understanding in vivo joint drug dynamics and genotypic resistance will progress at an optimum pace.