Skip to main content

Advertisement

Table 1 Study population characteristics by tumor AR expression

From: Prognostic and predictive value of androgen receptor expression in postmenopausal women with estrogen receptor-positive breast cancer: results from the Breast International Group Trial 1–98

Characteristic AR positivea
(n = 2475)
AR negativea
(n = 546)
P valueb
Treatment assignment, %
 Tamoxifen 21 21  
 Letrozole 22 18  
 Tamoxifen ➔ Letrozole 21 21  
 Letrozole ➔ Tamoxifen 21 20  
 Tamoxifen (2-arm period) 7 12  
 Letrozole (2-arm period) 8 10  
Received chemotherapy, % 29 32 0.21
Local therapy, % < 0.001
 Breast conserving surgery with radiation therapy 64 55  
 Breast conserving surgery without radiation therapy 4 5  
 Mastectomy with radiation therapy 12 18  
 Mastectomy without radiation therapy 20 22  
Age at randomization, mean (SD) 62 (8) 62 (8) 0.10
Tumor size, % < 0.001
 ≤ 2 cm 67 54  
 > 2–< 5 cm 29 41  
 ≥ 5 cm 3 4  
 Missing 1 1  
Number of lymph nodes positive, % 0.15
 0 63 58  
 1–3 26 30  
 4–9 7 8  
 10+ 3 3  
 Not evaluable 1 1  
Tumor grade, % < 0.001
 1 22 14  
 2 57 50  
 3 21 36  
 Missing 0 1  
ER expression (%), median (25th–75th percentile) 95 (90–99) 90 (75–95) < 0.001
PR expression (%), median (25th–75th percentile) 70 (20–90) 50 (5–84) < 0.001
 Missing, n 21 8  
Ki67 expression (%), median (25th–75th percentile) 12 (6–18) 12 (6–20) 0.07
 Missing, n 59 31  
HER2 expression, % 0.04
 Positive 5 8  
 Negative 95 92  
 Missing 0 0  
  1. aAR expression is defined as ≥ 1% positive, < 1% negative
  2. bHypothesis testing for differences in distributions between AR+ and AR− tumors used chi-square and Wilcoxon rank sum tests for categorical and continuous endpoints, respectively. Design variables (two- or four-arm randomization period and treatment assignment) were not compared. Individuals with missing data were excluded when performing hypothesis tests for continuous variables