11.07.2015 Views

Clinical Trials

Clinical Trials

Clinical Trials

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

❘❙❚■ Chapter 21 | Analysis of Survival DataIntroductionIn many clinical trials, the outcome is not just whether an event occurs, but alsothe time it takes for the event to occur. For example, in a cancer study comparingthe relative merits of surgery and chemotherapy treatments, the outcomemeasured could be the time from the start of therapy to the death of the subject.In this case the event of interest is death, but in other situations it might be theend of a period spent in remission from cancer spread, relief of symptoms, or afurther admission to hospital. These types of data are generally referred to astime-to-event data or survival data, even when the endpoint or the event beingstudied is something other than the death of a subject. The term survival analysisencompasses the methods and models for analyzing such data representing timefree from events of interest.Example: pancreatic cancer trialThe death rate from pancreatic cancer is amongst the highest of all cancers.A randomized controlled clinical trial was conducted on 36 patients diagnosedwith pancreatic cancer. The aim of this trial was to assess whether the use of a newtreatment A could increase the survival of patients compared to the standardtreatment B. Patients were followed-up for 48 months and the primary endpointwas the time, in months, from randomization to death. Table 1 displays thesurvival data for the 36 patients. We will use this example to illustrate somefundamental survival analysis methods and their applications.Basic concepts in survival analysisCensoringIn survival analysis, not all subjects are involved in the study for the same lengthof time due to censoring. This term denotes when information on the outcomestatus of a subject stops being available. This can be because the patient is lost tofollow-up (eg, they have moved away) or stops participating in the study, orbecause the end of the study observation period is reached without the subjecthaving an event. Censoring is a nearly universal feature of survival data. Table 2summarizes the main reasons for censoring that can occur in a clinical trial.Survival analysis takes into account censored data and, therefore, utilizes theinformation available from a clinical trial more fully.236

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!