Surrogate endpoints create potential, require statistical validation

Transforming statistical data into a surrogate endpoint can enable comprehension of trial results sooner than usual, but few make the cut after formal validation, says drug development expert and DIA panelist.


Interview of Dr. Marc Buyse, ScD, Chef Scientific Officer, IDDI  by Maggie Lynch, Reporter at

What are surrogate endpoints?

A surrogate endpoint is one that substitutes for a clinical endpoint during a clinical trial. They allow researchers to measure short-term outcomes such as tumour shrinkage or changes in biomarkers, rather than more long-term impacts of treatment, such as overall survival or improved quality of life.

A good surrogate endpoint is one that is statistically validated and can be relied upon to predict if the experimental treatment will reach the stated clinical endpoint.

How can surrogate endpoints accelerate approval?

Surrogate endpoints can be measured earlier than clinical endpoints, allowing for the results of a clinical trial to be understood sooner than would usually be possible. This can shorten the time it takes make new treatments available to patients, though it’s worth noting that full approval may be conditional on the longer-term follow-up of clinical endpoints such as event-free or overall survival.

How are new technologies enabling better understanding and use of surrogate endpoints?

We are learning more about the inner workings of the human body all the time, and medical knowledge has been rapidly accelerating in recent years.

New technologies, such as functional imaging and systems to detect circulating tumour DNA, can identify previously unknown biomarkers with the potential to become surrogate endpoints.

In which disease areas could surrogate endpoints be a critical factor in clinical trials and research?

Surrogate endpoints are critical in chronic and long-term conditions where clinical endpoints such as overall survival can take years to measure. They also play an important role in life-threatening diseases such as cancer, by helping to ensure timely access to innovative new treatments.

What makes a good surrogate endpoint?

A surrogate endpoint can only be used if there is extensive evidence that it can be relied upon to predict clinical benefit in the context of the trial. High blood pressure, for example, is an accepted surrogate for cardiovascular risk as the correlation is well understood.

Potential surrogates will usually require formal statistical validation in order to be acceptable from a regulatory point of view.  In fact, there is such a high bar for acceptance that very few surrogate endpoint candidates have met the grade.

IDDI has developed specific tools to meet this challenge and has partnered with multiple sponsors in this area.

Most recently, a team of IDDI experts worked with Roche to validate disease-free survival as a surrogate for overall survival in trials in HER-2-positive, early breast cancer. This analysis was conducted using methodology developed in part at IDDI, and also used by the FDA. Roche had selected IDDI thanks to the company’s unparalleled experience in the identification and validation of potential surrogate endpoints (more than 100 publications in the peer-reviewed literature).