The use and reporting of patient-reported outcomes in phase III breast cancer trials

Abstract

Background

Public and government attention to patient-centered research outcomes has been increasing, evidenced by the recent formation of the Patient Centered Outcomes Research Institute. Drug development clinical trials can be made more patient-centered by collecting patient-reported outcome measures that can inform decision making by patients and their health-care providers. Patient-reported outcomes are important to collect in trials of breast cancer therapeutics, which encompass a wide range of treatment regimens and side effects.

Purpose

We sought to determine recent trends in the use of patient-reported outcomes in drug trials for the treatment of breast cancer and evaluate the reporting of these data in study publications.

Methods

We searched ClinicalTrials.gov for phase III breast cancer drug trials, recording information on start date, primary completion date, primary outcome measure, primary sponsor, stage of cancer, and patient-reported outcome use. To assess the reporting of patient-reported outcome data, Google.com and PubMed.gov were searched for all publications resulting from included trials.

Results

We found 236 eligible trials, starting between May 1989 and December 2011. Of these trials, 83 (35%) stipulated patient-reported outcome use. The rate of patient-reported outcome use in recent years has shown no increase over earlier time periods: 37% (1989–2000) versus 36% (2004–2007) versus 30% (2008–2011) (p = 0.8). Trials with sponsorship led by the pharmaceutical industry and trials including patients with locally advanced or metastatic disease had the highest rates of patient-reported outcome use (40/87 (46%) and 44/102 (43%), respectively). Among the 83 trials that collected patient-reported outcome measures, 36 were completed a minimum of 2 years before our analysis; of these 36 studies, 19 (53%) had published patient-reported outcome data.

Limitations

Data were limited to self-reported descriptions of trials listed on the ClinicalTrial.gov database, which is the best compendium of trial information available, but it is neither a complete nor a fully accurate record of all trials.

Conclusions

Patient-reported outcome use and reporting in breast cancer drug trials has remained relatively low despite calls for more patient-centered research. Increasing the collection and availability of patient-reported outcome data to guide clinical decisions will require aligned support from trial sponsors, researchers, journal editors, regulators and patient advocacy groups, who can all play important roles in implementing change.