The study of tamoxifen and raloxifene (STAR): First report of patient-reported outcomes (PROs) from the NSABP P-2 Breast Cancer Prevention Study
LBA561 Background: STAR is a double-blind, randomized phase III prevention trial designed to evaluate the relative efficacy of raloxifene (R) compared to tamoxifen (T) in reducing the incidence of invasive breast cancer in high-risk postmenopausal women. In addition to standard safety monitoring, PROs were measured for quality of life (QOL) and symptoms (SXs). Methods: QOL was measured with the MOS SF-36, the CES-D, and the MOS Sexual Activity Questionnaire. SXs were measured using a modified checklist (SCL) from the NSABP P-1 BCPT. Primary endpoints were the SF-36 physical (PCS) and mental (MCS) component scales. A sample size of 1,670 provided statistical power of at least 0.8 (two-tailed test, alpha = .05). The QOL study was open to accrual at selected sites between 1/4/2000 and 5/31/2001, with questionnaires administered at baseline (pre-treatment), q 6 mos until 60 mos, and at 72 mos. SCL data were collected on all STAR participants. Longitudinal analyses used mixed effects modeling for the PCS, MCS, CES-D, and regression analyses were used to compare the average severity of SCL scores between T and R. Results: 19,747 participants enrolled in STAR and were eligible for the SCL assessment, with median follow-up time (mFU) 4.6 yrs. The QOL study enrolled 1,983 participants, 973 assigned to T and 1,010 assigned to R, mFU 5.4 yrs. QOL participants were comparable to women accrued concurrently at non-participating institutions. SCL and QOL forms compliance was high (with a low of 75% at 4.5 yrs for SCL and 74% for QOL). Mean PCS and MCS scores declined modestly over the 60 mos on study, with no significant difference between T and R. Mean CES-D scores worsened slightly on study, but with no significant difference between T and R. However, statistically significant differences (all p’s < .002) were noted between the T and R for severity of SXs, with R women reporting greater musculoskeletal problems, dyspareunia and weight gain and T women reporting greater vasomotor symptoms, leg cramps, bladder and gynecological problems. Conclusions: There were no significant differences between T and R in PROs for physical & mental health or depresson. While SX severity was generally low in this postmenopausal sample, the pattern of SXs differed between T and R. [Table: see text]