Randomized placebo-controlled clinical trials (RCTs) are the primary methodology for ascertaining the efficacy and safety of novel drugs, but their overall performance has been under scrutiny. This is especially true when novel putative antidepressants are assessed, a class of drugs that aims to treat Major Depressive Depression (MDD), one of the world’s most significant health problems. In early 2000, reports indicated that up to 50% of RCTs with antidepressants failed, regardless of whether the pharmacological agents under study were new or already on the market. The reasons were varied and have been studied extensively over the past 20 years, with the adoption of innovative designs, patient stratification procedures, and clinical trial simulations that have significantly improved the RCT performance. This research has led, among other results, to several attempts to better manage placebo response. In fact, since placebo response is the benchmark against which each active treatment should be statistically compared, the presence of strong and variable responses in the placebo study arm could “blur” the therapeutic response measured in the active study arm, that will not reach statistical significance, resulting in RCT failure. Identification of informative populations and implementation of sequential trial design were proposed to fight the excessive placebo response [1].

A commonly accepted approach to reduce placebo response is the inclusion of a lead-in (or run-in) period of several weeks in the RCT design, in which all recruited participants will blindly receive placebo, their initial response being monitored before randomization into the active and control arms. The assumption is that these placebo-responsive participants will be more likely to respond to placebo also during the dosing period of the RCT, inflating the outcome. Therefore, the exclusion of this placebo-prone population from the final analysis could benefit the signal detection of drug efficacy.

Log in or create a free account to read this content

Gain free access to this article, as well as selected content from this journal and more on nature.com

Access through your institution

or

Sign in or create an account
Continue with Google
Continue with ORCiD