Table 1.
Comparison of imaging characteristics between benign and malignant findings.
Fig 1.
Color-coded correlation matrix of Spearman’s rank correlation coefficients.
Blue indicates positive, whereas red indicates negative correlations. Size corresponds to strength of correlation. Correlation coefficient values are mirrored in the lower left quadrant. All values were statistically significant (p<0.001).
Table 2.
Odds ratios for the investigated covariates regarding presence of breast cancer.
Fig 2.
Odds ratios for presence of malignancy.
Odds ratios for malignant diagnosis (squares) are indicated along with the respective 95%-confidence intervals (lines) for BPE, FGT and age as determined by multivariate and univariate analysis. * marks significant odds ratios in favor of (>1) or against (<1) malignant diagnosis (p< = 0.001).
Table 3.
Cross-tabulation of age groups and BPE.
Table 4.
Cross-tabulation of BPE and hormonal receptor status of malignant findings.
Fig 3.
Tissue activation leads to breast cancer depending on tissue type.
Hormones lead to breast tissue activation that is reflected by BPE levels in MRI. Normal tissues exhibit intact repair mechanisms (i.e. apoptosis, cell cycle arrest) that upon activation keep malignant transformation at bay. In breast tissues that lack these repair mechanisms (i.e. due to genetic defects or to genetic alterations due to radiation exposure), tissue activation will lead to a chain reaction that eventually causes malignant transformation. We speculate that tissue activation in this multi-step carcinogenesis becomes at one point no longer dependent on hormonal activation. Until then, reduction of endogenous hormones by RRSO or through antihormonal treatment can stop or slow down tissue activation as reflected by lowered BPE levels [27–30].