endpoint

The FDA and the National Institutes of Health define an endpoint jointly, in a shared glossary written so that regulators and researchers mean the same thing by it: a precisely defined variable intended to reflect an outcome of interest that is statistically analysed to address a particular research question. A full definition names what is measured, when it is measured, with what instrument, and how repeated measurements in one person are combined.

A trial nominates one primary endpoint, the question it is designed and sized to answer, and states its secondary endpoints in advance. That ordering matters when reading results. A primary endpoint that succeeds is the trial's own verdict; a secondary endpoint is a genuine prespecified question, but it is answered with fewer participants and less statistical protection, so it points somewhere rather than settling it. An endpoint chosen as a stand-in for how a patient feels, functions or survives is a surrogate endpoint, and it has to earn that role with evidence.