Surrogate endpoints: a key concept in clinical epidemiology

Abstract

Surrogate endpoints are biomarkers or intermediate outcomes that are used as substitutes for clinical outcomes of interest, often to expedite research or decision-making. In contrast, patient-important (or patient-centered) outcomes are health outcomes that are of direct relevance and importance to patients themselves; clinical trials may have measured the impact of the intervention on other endpoints related to, but different from, those of primary importance to patients. This article aims to elaborate on the use and understanding of surrogate endpoints. There should be a well-understood and scientifically grounded relationship between the surrogate (replacement) and the patient-important (target) endpoint it is intended to represent. It should be biologically plausible that changes in the surrogate will consistently and predictably reflect changes in the patient-important endpoint. The surrogate endpoint should show a threshold effect, meaning that a specific change (or state) in the surrogate with an intervention (relative to the comparator) is associated with a predictable (change in the) patient-important outcome. This helps establish a meaningful cutoff or target for the treatment effect on the surrogate endpoint. While surrogate endpoints offer advantages in certain situations, it is important to remember that their use requires careful validation to ensure they reliably predict the true clinical outcome. The validity of "surrogate endpoints" should be supported by robust scientific evidence and rigorous evaluation before these can be considered and labeled as surrogate endpoints.

Keywords: Clinical epidemiology; Endpoints; Indirectness; Outcome measures; Randomized controlled trials; Surrogate outcomes.