Abstract
Background
To further investigate the role of sex hormones in breast cancer, we assessed the relations of circulating estradiol and testosterone to tumor size and estrogen receptor (ER) status.
Method
This was a cross-sectional study including 492 postmenopausal breast cancer patients. The relation of circulating hormones to patient and tumor characteristics was assessed using the Fisher or Cuzick tests. Multivariable logistic regression was used to estimate the odds ratios (ORs) of having tumors ≥2 cm (vs <2 cm) and having ER-positive tumors (vs ER-negative) with increasing quartiles of estradiol and testosterone.
Results
Mean estradiol and testosterone levels increased significantly with increasing tumor size. The ORs of tumors ≥2 cm increased significantly with increasing quartiles of estradiol (Ptrend<0.001) and testosterone (Ptrend=0.005). When adjusted for estradiol, the association between testosterone and tumor size was no longer significant. Mean testosterone levels were higher in ER-positive than ER-negative patients (P<0.001), while mean estradiol levels did not differ significantly between the two ER categories (p=0.192). The ORs of having an ER-positive tumor increased significantly with increasing quartiles of testosterone (Ptrend=0.002), whereas the increase with increasing estradiol quartiles was not significant (Ptrend=0.07).
Conclusion
The association of both hormones with tumor size implies that both are involved in tumor growth, testosterone mainly by conversion to estradiol. The strong association of testosterone with ER contrasts with the weak association of estradiol with ER and confirms testosterone as a marker of hormone-dependent tumors. These findings suggest that testosterone evaluation might be useful to better identify patients with hormone-dependent disease.