Sequential analysis study design

Figure 11.2 Illustration of the Armitage sequential analysis study design. Patients are paired, and one of each pair receives each alternative treatment. If the patient receiving treatment A does better than the one receiving treatment B (A > B), then the line moves upwards vice versa, if the patient receiving B does better than the one receiving A (B > A), then the line moves downward. If the treatments cannot be distinguished within a pair of patients, then the line moves horizontally. The critical boundaries (broken lines) are computed from prospective measures of a and fj (e.g. p = 0.05 and 80% power, respectively). The technique derives from an engineering control chart and, once again, can be adapted to more sophisticated forms, including limits on the study size for indeterminate results...

The study uses a design that permits a valid comparison with a control to provide a quantitative assessment of drug effect. The protocol for the study and report of results should describe the study design precisely for example, duration of treatment periods, whether treatments are parallel, sequential, or crossover, and whether the sample size is predetermined or based upon some interim analysis. Generally, the following types of control are recognized ... 

Meta-analysis, Group Sequential Study Designs, Centralized Endpoint Adjudication, and Composite Endpoints... 

In our study we compare two diversity-driven design methods (uniform cell coverage and clustering), two analysis methods motivated by similarity (cell-based analysis and cluster-classification), and two descriptor sets (BCUT and constitutional). Thus, our study addresses some of the many questions arising in a sequential screen how to choose the initial screen, how to analyze the structure-activity data, and what molecular descriptor set to use. The study is limited to one assay and thus cannot be definitive, but it at least provides preliminary insights and reveals some trends. 

We decided to examine only one sample size for the design, 4096. Abt et al. (6) examined sample size, among other factors, when studying sequential screening. Yi et al. (20) also used a sample size of 4096 when studying the optimization of a statistical analysis method for this dataset. Both studies indicated that relatively small sample sizes of 5000 to 10,000 compounds could be used to produce useful trees. Clearly, large sample sizes should lead to better... 

Toxicokinetics studies are designed to measure the amount and rate of the absorption, distribution, metabolism, and excretion of a xenobiotic. These data are used to construct predictive mathematical models so that the distribution and excretion of other doses can be simulated. Such studies are carried out using radiolabeled compounds to facilitate measurement and total recovery of the administered dose. This can be done entirely in vivo by measuring levels in blood, expired air, feces, and urine these procedures can be done relatively noninvasively and continuously in the same animal. Tissue levels can be measured by sequential killing and analysis of organ levels. It is important to measure not only the compound administered but also its metabolites, because simple radioactivity counting does not differentiate among them. 

Clinical trials, with almost no exception, are longitudinal (Chow and Liu, 2004). This means that data are accumulated sequentially over time. From the perspective outlined so far in the book, the statistical analysis takes place once the number of subjects stated in the study protocol have been enrolled, randomized, and completed their participation in the trial. This approach can be called the Fixed design or fixed sample design approach. Another design of interest in clinical trials is the group sequential design, in which interim analysis plays a crucial role. 

The analysis of a supersaturated design is usually conducted by using some type of sequential model-fitting procedure, such as stepwise regression. Abraham et al. (1999) and Holcomb et al. (2003) have studied the performance of analysis methods for supersaturated designs. Techniques such as stepwise model fitting and all-possible-regressions type methods may not always produce consistent and reliable results. Holcomb et al. (2003) showed that the performance of an analysis technique in terms of its type I and type II error rate can depend on several factors,... 

---