Surrogate endpoints need complementary patient reported outcomes

Cancer’s gold standard trial endpoint, overall survival (OS), is being overtaken by the growing use of surrogate endpoints. This presents regulators with challenges, but these are seldom debated in patient circles and we ignore the fact that surrogate endpoints focus on drug effect rather than patient benefit. For many patients, the recent article published in The BMJ by Dawoud and colleagues, putting forward proposals for rigorous systematic validation of surrogate endpoints, will be seen purely as a technical paper.

I feel a bit raw over a recent experience of a study with a surrogate primary endpoint. The use of oloratumab (combined with doxorubicin) as first line treatment for advanced metastatic soft tissue sarcoma gained interim approval by the European Medicines Agency (EMA) and the National Institute for Health and Clinical Excellence (NICE) based on an exceptional OS result in a phase II study where OS was a secondary endpoint. [1] The primary progression free survival endpoint did not merit that decision. When early results from the phase III study were announced some months later, however, the OS result was not confirmed; the treatment was, quite properly, withdrawn by the manufacturer. [2] I was not involved in these studies, but like most sarcoma patients who saw the phase II result, I was supportive of getting the treatment into use. Some clinicians were doubtful, justifiably as it turned out.  We are still learning from our misplaced enthusiasm.

Medicines regulators, such as the US Food and Drugs Administration (FDA), the European Medicines Agency (EMA), and the Medicines and Healthcare Products Regulatory Agency (MHRA), and health technology appraisal (HTA) regulators (such as NICE) ask different questions. The former consider safety and clinical efficacy while the latter look at cost effectiveness—the balance between patient benefit and the impact on the healthcare system. They look at data in different ways to get the answers they need. One type of data that may help tackle the shortcomings of surrogate endpoints are patient reported outcomes (PROs). These metrics go beyond “quality of life” and look at specific impacts of disease or side effects of treatment, physical and social function, and psychological impact. They build a picture of the condition from a patient perspective.

And while the pharmaceutical world is showing some enthusiasm for PROs, the HTA world still isn’t. Many of us find it hard to understand why. 

Patients want clinical trials to resolve uncertainty. We recognise that questions need to be asked, we look to clinical researchers to deliver the answers. We are keen to help make that happen appropriately and effectively. Patient demand for access to new drugs, shorter timescales, and smaller trial cohorts, alongside company needs to generate a return from the ever increasing costs of drug development, are easy to comprehend. Measurement of effect, inclusion criteria, dosing, and side effects are matters we work to understand and we welcome the growing use of PROs to provide subjective endpoints which complement the objective medical measures. 

The problem is that PROs are still an evolving science. It seems they can offer valuable information to the medicines regulator, while the world of HTA identifies systematic weaknesses which mean PROs come into consideration only when other evidence is weak and more flexible thinking is required.

Surrogate endpoints are here to stay, but patients want to see PROs complementing them as co-primary or even sole primary endpoints. [3] The case for this duality, subjective alongside the objective, seems inarguable. Academic collaborations such as the Spirit-Pro Extension,  which aims to identify best practice for developing interventional studies, and Sisaqol, led by the European Organisation for the Research and Treatment of Cancer, are establishing standards for analysis. [4,5]

We need to balance two differing viewpoints, a medical story and a parallel patient story. They rarely differ by very much, but any difference could be important.

A more rigorous validation of surrogate endpoints will create some clarity and may encourage more use of PROs. There have been some important contributions to PRO development by medicine regulators, and both the EMA and FDA have published guidance on their use in cancer. [6,7] What we now need to see is stronger involvement by the HTA regulators in the development of PROs, to bring them into mainstream thinking.

Even though regulators do not lead or sponsor studies, patients want to see them contributing positively to the evolution of PROs. The benefit will be that patients will understand trial results better, while the doctors guiding us to treatment decisions will have less uncertainty and stronger evidence to use.

Roger Wilson is a long term recurrent and metastatic soft tissue sarcoma survivor. He is a member of the NCRI Consumer Forum and founder of Sarcoma UK.

Competing interests: none declared.


1.Tap WD, Jones RL, Van Tine BA, et al Olaratumab and doxorubicin versus doxorubicin alone for treatment of soft-tissue sarcoma: an open-label phase 1b and randomised phase 2 trial.  Lancet. 2016 Jul 30;388(10043):488-97.

2. Lilly reports results of phase 3 soft tissue sarcoma study of Lartruvo. 18 January 2019.

3.Wilson R, Patient led PROMs must take centre stage in cancer research. Research Involvement and Engagement. 2018.

4. Calvert M, Kyte D, Mercieca-Bebber R, et al., Guidelines for Inclusion of Patient-Reported Outcomes in Clinical Trial Protocols The SPIRIT-PRO Extension.  JAMA. 2018;319(5):483-94.

5. Bottomley A, Pe M, Sloan J, et al. Analysing data from patient-reported outcome and quality of life endpoints for cancer clinical trials: a start in setting international standards. Lancet Oncol. 2016;17(11):e510-e514. 

6. EMA. Appendix 2 to the guideline on the evaluation of anticancer medicinal products in man – The use of patient-reported outcome (PRO) measures in oncology studies.

7. FDA. Guidance for industry patient-reported outcome measures: use in medical product development to support labelling claims.