Search

As part of the evaluation of phase II trials, it is common practice to perform exploratory subgroup analyses with the aim of identifying patient populations with a beneficial treatment effect. When investigating targeted therapies, these subgroups are typically defined by biomarkers. Promising results may lead to the decision to select the respective subgroup as the target population for a subsequent phase III trial. However, a selection based on a large observed treatment effect may potentially induce an upwards-bias leading to over-optimistic expectations on the success probability of the phase III trial. We describe how Approximate Bayesian Computation techniques can be used to derive a simulation-based bias adjustment method in this situation. Recommendations for the implementation of the approach are given. Simulation studies show that the proposed method reduces bias substantially compared with the maximum likelihood estimator. The procedure is illustrated with data from an oncology trial. Copyright © 2017 John Wiley & Sons, Ltd.

The primary goal of phase II studies is to assess the efficacy of the new treatment in order to decide whether it has sufficient activity to warrant further evaluation in a phase III comparative trial. However, many adequately conducted phase II trials are negative leading to termination of drug development. Heterogeneity of the population is often considered to be a cause of treatment effect dilution. One approach to determine the sensitive subpopulation is to conduct several phase II trials, one in each specific subset of patients. This option might unethically increase the number of non-sensitive patients under evaluation. Adaptive two-stage designs have been recently proposed. London and Chang proposed a global one-sample test for response rates for stratified phase II clinical trials, whereas Jones and Holmgren proposed an adaptive design that allows preliminary determination of efficacy that may be restricted to a specific subpopulation defined by biomarker status. These two methods do not allow early termination for efficacy in one or several subgroups as they are extensions of the Simon design. The authors propose an alternative method to deal with stratification in phase II clinical trials and identification of the best target population. This method is based on the multiple-stage Fleming design allowing for early stopping rules for either efficacy or inefficacy. It also integrates a procedure testing whether treatment effects are similar or heterogeneous between the two groups. The operating characteristics of this method were compared with those of a standard Fleming design using exact binomial probabilities. Copyright © 2011 John Wiley & Sons, Ltd.

Over the years, concerns have been raised regarding the appropriateness of using the average bioequivalence approach for evaluation of comparability between formulations. In lieu of average bioequivalence, scientists from academia, industry and regulatory agencies have spent considerable effort and time in exploring the concepts of population and individual bioequivalence, and developing the statistical methods to assess the bioavailability metrics using these approaches. Recently, the Food and Drug Administration (FDA) has published a preliminary draft guidance entitled 'In vivo bioequivalence studies based on population and individual bioequivalence approaches'. The concept of prescribability and switchability underscores the difference between the population and individual bioequivalence approaches. The most important consideration for individual bioequivalence, the focus of this paper, rests on the assurance that products deemed bioequivalent can be used interchangeably in the target population (switchability). In addition to the comparison of averages, the individual bioequivalence approach compares within-subject variabilities and assesses subject-by-formulation interaction. The proposed criterion represents substantial departure from the current practice and thus has resulted in extensive public discussion. In contrast to the current average bioequivalence procedure, the proposed individual bioequivalence approach offers flexible equivalence criteria based on the individual therapeutic window and variability of the reference drug product. The proposed criterion rewards manufacture of less variable drug products, allows scaling criteria for highly variable/narrow therapeutic range drugs, and promotes the use of subjects from the general population in bioequivalence studies. The FDA is currently considering various approaches for resolution of issues raised from the public debate on the subject-by-formulation interaction term, statistical methods and resource implications.

This paper examines the amount of variation among centres and estimates the overall effect of therapy in a multi-centre cancer clinical trial with censored failure time data. To investigate the centre effects, the variation in the treatment effect must be taken into consideration in addition to the variation in the baseline risk. We treat centre effects as random ones and extend the penalized partial likelihood approach proposed by McGilchrist to estimate the treatment-by-centre interaction as well as the baseline risk. This method is applied to data from a superficial bladder cancer clinical trial investigating the efficacy of intravesical chemotherapy after transurethral resection. In this trial, although there exists some degree of centre variation, especially in the baseline risk, the treatment is effective in preventing recurrence among the participating centres. This result indicates that the treatment effect is generalizable to the target population.

Non-invasive ambulatory blood pressure monitoring is an appropriate technique for the identification of the target population requiring antihypertensive treatment, since the systematic bias seen in casual clinic measurement is removed. The use of the technique in controlled clinical trials leads to a reduction in the sample size required to show a treatment difference in antihypertensive effect compared to sporadic single measurements, although in practice the number of recruited patients can considerably exceed the number of evaluable patients, due to poor compliance or incompleteness of data. In addition, the equality of observation essential in clinical trials cannot be controlled to the same degree as is possible with clinic measurements. Comparison of estimators of the overall characteristic blood pressure reveals that the 24-hour mean value produces an adequate estimator. A description of the individual blood pressure profile is obtained using smoothing procedures, the moving interval average providing a robust and simple approach. Clinically meaningful characteristics of the profile are difficult to define due to a dearth of information on prognostic importance of the features of the 24-hour blood pressure profile. Smoothed individual profiles can be combined to produce graphical descriptive treatment group summaries.

We provide an account of a study to assess reference limits for eight routine laboratory determinations at the Academic Hospital of the Free University, Amsterdam and emphasize methodological issues rather than results. We argue that reference limits have use mainly in the first phase of the diagnostic process. Reference and target populations should be grossly comparable, and therefore patients (after slight selection) could well serve as references. However, we found major differences between in- and out-patients, so we suggest that this factor, together with age and sex, be taken into account. To arrive at reliable limits, the size of the reference sample should be at least 100. Laboratory reports should provide percentiles, which enable a more flexible decision than do fixed limits.

A considerable number of the clinical reports which are presented to the Dutch Board for the Evaluation of Drugs, have deficiencies and/or shortcomings. A number of these, including loose description of the target population and sampling method, methodological flaws, incorrect treatment of withdrawals, confounding of patients and observations per patient, incomplete or incorrect description of the data and incorrect or inappropriate use of statistical tests, are discussed.