In the above referenced article, we inadvertently and consistently made two typographical errors. First, we switched the genotyping assignments in our tables; that is, for the -401 T/C polymorphism, the allele we reported as the “T” allele should have been labeled “C”, and for the -354 A/G polymorphism, the allele reported as the “A” allele should have been labeled “G”. Accordingly, the true findings are the reverse of what are reported in the paper. A corrected version of the main results table (Table 4) appear below. Second, we switched the assignment of genotypes reported by previous studies1, 2, 3, 4, 5 Because we reversed both our data and the data reported by others, the comparison of our results with previous work remains unchanged. Specifically, our findings are consistent with most previous studies. Two studies have reported a significant decrease in risk associated with carriage of the G allele and no association with the -401 T/C polymorphism.1, 4 The Shanghai Breast Cancer Study also reported a decrease in risk among G/G homozygotes, although the results were not significant.2 However, that study did find a significant increased risk of breast cancer associated with carriage of a T allele: compared to C/C carriers, age-adjusted odds ratios (95% CI's) for T/C and T/T were 1.3 (1.0, 1.7) and 1.4 (1.1, 1.8), respectively. The -401 T allele has been associated with younger onset of disease3, 5 and the differences in age distribution between our study cohort and the Shanghai cohort may explain our different findings.

Table IV. Estimated Relative Risk of Breast Cancer by Factor and Genotype in the Study of Osteoporotic Fractures, 1986-2003*
FactorESR1 ISV1 -401ESR1 ISV1 -354
  • *

    Proportional hazards regression models. Data were controlled for study site; age, age at menopause, parity, body mass index, and bone mineral density as continuous variable, and; parous, surgical menopause and ever smoked as dichotomous variables (n/y), execept in those models stratified by a particular variable.

  • **

    Significant at p < 0.005.

BMI (kg/m2)
< median (26.2)11.061.32 10.891.06 
≥ median10.990.760.3210.970.670.46
Age at Menarche
< 1310.970.72 11.0070.47 
≥ 1311.171.230.3710.861.020.21
Age at Menopause
< median11.121.04 10.811.05 
≥ median10.940.930.8911.020.650.38
yes10.60.58 10.670.73 
Age at First Birth
< 3511.191.16 11.0040.85 
≥ 3511.461.180.1111.040.680.30
Current HT Use
no10.930.94 10.890.85 
Ever HT Use
no11.050.97 10.990.74 
Ever Smoke
no10.980.86 10.740.60 
Current Drinker
no10.870.79 10.890.71 
Walks for Exercise
no11.270.89 11.180.49 
yes10.861.090.1610.751.150.01 **
Family History of Breast Cancer
no11.071.11 10.950.89