Overall survival of participants compared to non-participants in a randomized-controlled trial (SELECT BC): A prospective cohort study.

Authors:

Abstract
2527 Background: Whether participation in a randomized-controlled trial provides benefits to patients is controversial. Some studies have showed that participants have longer prognosis than non-participants. However, in such studies, underlying bias always exists, such as patient condition, patient selection by physicians, and differences in the care provided. Therefore, to minimize such bias, we enrolled non-participants at the time of patient recruitment for a randomized-controlled trial, and prospectively compared the prognoses of participants and non-participants. Methods: This was a prospective cohort study accompanying the “SELection of Effective ChemoTherapy for Breast Cancer” (SELECT BC) study, which compared S-1 and taxane as first-line treatment for metastatic breast cancer and reported the non-inferiority of S-1 with respect to overall survival. Prognoses of trial participants and non-participants were compared. The endpoint was overall survival [Lancet Oncol 17:90-98, 2016]. All patients met the eligibility criteria set for SELECT BC. Results: From August 2009, when this study started, to July 2010, 228 patients were enrolled in SELECT BC, while 65 patients were registered as non-participants in this study. The median follow-up period was 32.5 months. The median overall survival of participants (36.8 months) was longer than that of non-participants (25.2 months) (hazard ratio [HR]: 1.48; 95% confidential interval [CI]: 1.06-2.07; p = 0.023). Standard chemotherapy was administered to 48 non-participants (74%); 34 (52%) of them received standard taxane regimen or S-1, which was to be administered if they participated in the trial. However, the median overall survival of these 34 patients (22.0 months) was shorter than that of the trial participants (36.8 months) (HR: 2.03; 95% CI: 1.35-3.06; p = 0.0007). Conclusions: Randomized-controlled trial participants had longer overall survival than non-participants. Several factors, such as mental status, relationship between physicians and patients, or difficulties in receiving the recommended treatment could have influenced the prognosis. Clinical trial information: 000020645.