Clinical Endpoint

A clinical endpoint is a prespecified, measurable outcome used to evaluate the effect of an intervention in a trial. The primary endpoint is the main outcome on which the trial’s sample size and principal conclusion are based; secondary endpoints address additional questions and are interpreted with more caution. Endpoints may be hard clinical outcomes that directly reflect how patients feel, function or survive, such as death, myocardial infarction or stroke, or surrogate endpoints, such as HbA1c, blood pressure or body weight, which are biomarkers expected to predict clinical benefit. Composite endpoints combine several events into one outcome, such as major adverse cardiovascular events (MACE).

STEP 1 used percentage change in body weight and achieving at least 5% weight reduction as coprimary endpoints. The SELECT trial instead used a hard composite endpoint: death from cardiovascular causes, nonfatal myocardial infarction or nonfatal stroke. Among 17,604 people with cardiovascular disease and overweight or obesity but no diabetes, this endpoint occurred in 6.5% on semaglutide and 8.0% on placebo over a mean follow-up of 39.8 months (hazard ratio 0.80).

The endpoint defines what a trial can actually show. A drug that lowers a lab value has met a surrogate endpoint, but that is not the same as proving it helps people live longer or avoid illness. Fleming and DeMets described treatments that improved surrogate markers yet failed to improve, or even worsened, clinical outcomes, such as antiarrhythmic drugs that suppressed abnormal heart rhythms but increased mortality. Surrogate results need confirmation against outcomes patients experience.

Surrogate endpoints are still useful. Outcomes such as body weight or HbA1c change within months, so trials can be smaller and shorter than those waiting for heart attacks or deaths. STEP 1 and SURMOUNT-1, for example, used body weight change as a coprimary endpoint. Hard-outcome trials then test whether the benefit carries through: SELECT showed that semaglutide reduced major cardiovascular events in people with obesity and cardiovascular disease but without diabetes, a different question from weight loss alone.

The primary endpoint should be fixed before the trial starts, along with the analysis plan. When many secondary endpoints are measured, some will reach statistical significance by chance, so a trial that misses its primary endpoint but highlights a positive secondary one should be read cautiously. When reading claims about any peptide, ask what was measured, whether it was the prespecified primary endpoint, and whether it is a surrogate or a real clinical outcome.

“If a drug improves a biomarker, it improves health”

Biomarkers are only useful when changes in them reliably predict changes in clinical outcomes, and that link can break. Treatments have improved surrogate measures while failing to improve, or even worsening, survival or other hard outcomes. This is why regulators and clinicians value trials that measure clinical events directly.

“All endpoints in a trial carry equal weight”

The primary endpoint is the one the trial was designed and powered to test. Secondary and exploratory endpoints can generate useful hypotheses, but they are more vulnerable to chance findings, especially when many are analyzed. A positive secondary result does not rescue a trial that missed its primary endpoint.

“A composite endpoint means every component improved”

A composite such as MACE combines several events, and an overall benefit can be driven mostly by one of them. Readers should look at the individual components reported in the trial rather than assuming each one changed by the same amount.

What is the difference between primary and secondary endpoints?

The primary endpoint is the main outcome a trial is designed to answer; the sample size and the principal conclusion depend on it. Secondary endpoints are additional outcomes that add context, such as quality of life or other risk factors. Because many secondary endpoints may be tested, their results are usually treated as supportive rather than definitive.

What is a surrogate endpoint?

A surrogate endpoint is a laboratory measure or physical sign used in place of a clinical outcome, because it changes sooner and is easier to measure. Examples include HbA1c for diabetes, blood pressure, cholesterol and body weight. A surrogate is only trustworthy when it has been shown to predict the clinical outcome it stands in for.

What endpoint did the SELECT trial use?

SELECT used a composite primary endpoint of death from cardiovascular causes, nonfatal myocardial infarction or nonfatal stroke, analyzed as time to the first event. It enrolled 17,604 people aged 45 or older with established cardiovascular disease and overweight or obesity but no diabetes, comparing semaglutide with placebo.