IN THIS ISSUE

Extract

When Are “Positive” Trials Really Positive?

Clinical trials are declared “positive” when the treatment and control groups show a statistically significant difference in primary endpoints (e.g., overall survival, progression-free survival). Drugs are approved by the United States Food and Drug Administration and the European Medicines Agency based on such “positive” results, but often do not provide a clinically meaningful benefit to cancer patients. In this commentary, Ocana and Tannock (p. 16 ) suggest that to declare a clinical trial “positive,” investigators should also reach (or exceed) a difference in outcome that was prespecified in the protocol – a difference that reflects survival benefit or improved quality of life of patients.

In an editorial, Lee (p. 2 ) agrees with the authors that positivity of a trial should not be based on statistically significant P values alone. He mentions that even though the authors have taken a step in the right direction, comparing the observed effect on the treatment and control groups with a prespecified effect still may not be sufficient because of limited statistical power and other limitations. He draws attention to the superiority of Bayesian methods and their application to a more accurate assessment of “positivity” of clinical trials.