br Breast cancer subtype distribution among

3.2. Breast cancer subtype distribution among BPE types
Among all patients with marked BPE, triple negative subtypes were significantly higher (p = 0.0074).
3.3. Multivariate regression analysis
Fig. 1. Subtracted THRIVE Galactose 1-phosphate  enhanced MR image showing a Luminal B HER2- breast cancer of the right breast in a 54-year-old woman with mild BPE evaluated in the contralateral breast.  European Journal of Radiology 113 (2019) 148–152
Table 1
Data distribution by comparing BPE patterns with molecular breast cancer subtypes.
BPE LUMINAL A LUMINAL LUMINAL NON TRIPLE
B HER2- B HER2+ LUMINAL NEGATIVE
Fig. 2. Subtracted THRIVE contrast enhanced MR image showing a triple ne-gative breast cancer of the right breast in a 49-year-old woman with marked BPE evaluated in the contralateral breast.
Fig. 3. Breast cancer subtype distribution among different BPE patterns.
Multivariate regression analysis found no statistical significant ef-fect of age, menopausal status and lesion diameter on BPE distribution (p = 0.39; coefficient of determination R2 = 0.03; multiple correlation coefficient = 0.19). The following values were obtained for the single independent variables: age (p = 0.26; coefficient= -0.01); menopausal status (p = 0.16; coefficient = 0.3); lesion diameter (p = 0.98; coeffi-cient= -0.0006).
3.4. Logistic regression
A dichotomized approach on BPE with logistic regression was per-formed in order to verify the obtained data and the study sample was divided into two groups: group 1 (BPE 0; n = 44; 53.6%); group 2 (BPE1; n = 38; 46.3%). Logistic regression found no statistical sig-nificant value (p = 0.49). The following values were obtained for the single variables: age (p = 0.43; coefficient= -0.02); menopausal status
G. Dilorenzo, et al.
3.5. Inter-rater agreement
The inter-rater agreement in evaluating BPE patterns on MRI was almost perfect with Cohen’s k = 0.83.
4. Discussion
The introduction of MRI in the field of breast imaging has increased the sensitivity and specificity for detecting breast cancer respectively to 95%–99% and 80%. MRI also provides crucial information concerning the normal breast tissue features, including BPE, defined as the en-hancement of the normal breast tissue after contrast material injection. BPE can be classified as minimal (< 25% of glandular tissue demon-strating enhancement), mild (25%–50% enhancement), moderate (50%–75% enhancement), or marked (> 75% enhancement). Typically BPE is minimal or mild and shows diffuse, bilateral and symmetric distribution. In a minor percentage of patients, it could be moderate or marked, and could have asymmetric or not diffuse pattern, affecting breast cancer detection [1–4].
BPE is influenced by the anatomy of the breast’s arterial and venous systems, and by endogenous hormones levels during the menstrual cycle; therefore, it is recommended to perform breast MR in the first half of the menstrual cycle, in order to minimize the enhancement of the normal breast tissue. A negative BPE correlation with menopausal status has been reported [1,5–9]. Also hormone therapies can reduce BPE, like selective estrogen receptor modulators (SERMs) or aromatase inhibitors [10]. Verardi et al. [39] reported an association between asymmetric increased breast vascularity and ipsilateral cancer, parti-cularly for invasive cancers of more than 2 cm in diameter or with high pathologic grade. This data suggests to perform any evaluation of BPE on the contralateral breast. In fact, BPE was evaluated in the con-tralateral breast in all patients enrolled in our study.
Several studies investigated the effect of BPE on breast cancer de-tection and its association with different tumor types. Some authors found no correlation between BPE and mammographic density [11–14], but many studies demonstrated that increased BPE at breast MRI was associated with greatly increased risk of breast cancer in high risk patients [7,13,15] and with lower recurrence-free survival in breast cancer patients treated with neoadjuvant chemotherapy [16].
Particularly, Dontchos et al. and King et al. found a positive BPE correlation with breast cancer risk in high risk patients while Bennani-Baiti et al. found no correlation with breast cancer risk in non high risk patients [7,15,19]. Van der Velden et al. also stated that BPE in the contralateral breast of patients with invasive unilateral breast cancer was significantly associated with long-term outcome particularly in case of estrogen receptor-positive, human epidermal growth factor re-ceptor 2 (HER2)-negative breast cancer [17].
In fact, most relevant studies about breast cancer during last dec-ades involve its molecular characterization, using genetic array testing (or approximations to this method with immunohistochemistry) to detect the expression of hormonal and epidermal growth factor 2 re-ceptors in the tumor cells. This data, associated with Ki67 labeling index, allows to classify several cancer subtypes, which differ for etiology, natural history and response to treatment [21].