Improvements noted in area under the curve for both Gail and Rosner risk prediction models
WEDNESDAY, Oct. 30 (HealthDay News) -- The addition of circulating hormone levels correlates with improved breast cancer risk prediction, according to a study presented at the American Association for Cancer Research's International Conference on Frontiers in Cancer Prevention Research, held from Oct. 27 to 30 in National Harbor, Md.
Shelley S. Tworoger, Ph.D., from Brigham and Women's Hospital in Boston, and colleagues used data from 473 cases of postmenopausal invasive breast cancer and 770 controls. The subset of hormones most associated with risk was identified and improvement in the area under the receiver operating characteristic curve (AUC) was assessed, using the Gail and Rosner risk prediction models.
The researchers found that estrone sulfate, testosterone, and prolactin correlated with a 8.6-point increase on the AUC in the Gail model and a 3.8-unit increase for the Rosner score in the training dataset (70 percent of data), and with changes of 5.0 and 5.1 units, respectively, in the independent dataset (30 percent of data). Using a subset including estrone sulfate, testosterone, prolactin, and sex-hormone binding globulin, the changes in AUC were 7.3 and 5.5 for the Gail and Rosner scores, respectively, in the training dataset and 12.5 and 8.3, respectively, in the independent dataset.
"The improvement in prediction when adding circulating hormone levels was better than the improvement observed by other studies that included mammographic density and genetic factors," Tworoger said in a statement.
Press Release (http://aacrnews.wordpress.com/2013/10/29/measuring-hormones-could-help-improve-breast-cancer-risk-prediction/ )More Information (http://www.aacr.org/home/scientists/meetings--workshops/frontiers-in-cancer-prevention-research.aspx )