A clinical trial ¹ is any form of planned experimental study designed, in general, to evaluate the effect of a new treatment on a clinical outcome in humans. Clinical trials may either be pre-clinical studies, small clinical studies to investigate effect and safety (Phase I/II trials) or full evaluations of the new treatment (Phase III trials). In this chapter we discuss the main aspects of Phase III trials, all of which should be reported in any publication (see the CONSORT² Statement checklist in Table 14.1, and Figs 14.1 and 14.2).

Table 14.1 Checklist of items from the CONSORT Statement (Consolidation of Standards for Reporting Trials) to include when reporting a randomized trial.

Figure 14.1 The CONSORT Statement’s trial profile of a randomized controlled trial’s progress.

Figure 14.2 Trial profile example

(adapted from the trial described in Chapter 40 with permission).

Treatment Comparisons

Clinical trials are prospective studies in that we are interested in measuring the impact of a treatment given now on a future possible outcome. In general, clinical trials evaluate new interventions (e.g. type or dose of drug, or surgical procedure). Throughout this chapter we assume, for simplicity, that only one new treatment is being evaluated in a trial.

An important feature of a clinical trial is that it should be comparative (Chapter 12). Without a control treatment, it is impossible to be sure that any response is due solely to the effect of the treatment, and the importance of the new treatment can be overstated. The control may be the standard treatment (a positive control) or, if one does not exist, it may be a negative control, which can be a placebo (a treatment which looks and tastes like the new drug but which does not contain any active compound) or the absence of treatment if ethical considerations permit.

Primary and Secondary Endpoints

When choosing the endpoint at the planning stage of a study (Chapter 13), we must decide which outcome most accurately reflects the benefit of the new therapy. This is known as the primary endpoint of the study and usually relates to treatment efficacy. Secondary endpoints, which often relate to toxicity, are of interest and should also be considered at the outset. Generally, all these endpoints are analysed at the end of the study. However, we may wish to carry out some pre-planned interim analyses (for example, to ensure that no major toxicities have occurred requiring the trial to be stopped). Care should be taken when comparing treatments at these times owing to the problems of multiple hypothesis testing (Chapter 18). An independent Data Safety and Monitoring Committee (DSMC) often takes responsibility for the interpretation of interim analyses, the results of which should generally be treated confidentially and not circulated to other trial investigators unless the trial is stopped.

Subgroup Analyses

There is often a temptation to assess the effect of a new treatment in various subgroups of patients in the trial (e.g. in men and women separately; in older and younger individuals). Owing to the problems with multiple hypothesis testing and reduced study power (Chapter 18), these should be avoided unless they are pre-planned, the study sample size has been calculated accordingly, and appropriate statistical methods have been used for analysis.

Stay updated, free articles. Join our Telegram channel

Join