To the Editor A valid surrogate end point is a substitute for a true end point that allows conclusions about the effect of treatment on the true end point to be drawn sooner than with a true end point. Royce et al1 argued that a prostate-specific antigen nadir value greater than 0.5 ng/mL is a good surrogate end point for all-cause mortality in men with prostate cancer. They base their claim on the proportion of treatment effect explained (PTE) of 103.86%. However, PTE is not a reliable measure of surrogacy. First, the confidence intervals, which were not reported, are typically too large to be useful,2 particularly if the effect of treatment on the true end point (not adjusted for the surrogate end point) is small.3 Second, PTE is based on an assumption, which may not hold, that treatment and surrogate end point have an additive effect, with no interaction, on the hazard function for all-cause mortality.3
Baker SG. Validity of Surrogate End Points for Prostate Cancer. JAMA Oncol. 2018;4(1):129–130. doi:10.1001/jamaoncol.2017.0334
Customize your JAMA Network experience by selecting one or more topics from the list below.