Introduction

Although the relationship between proliferative breast lesions (PBLs) and breast cancer (BC) has been discussed, PBLs are known as an important risk group in BC development. The risk of BC increases according to the type of benign breast lesions. While there is no risk of BC in non-PBLs, this risk doubles on an average for PBL-without atypia (PBL-WOA) patients and increases by 4–6 times in female PBL-with atypia (PBL-WA) patients. Although several studies have been performed on the classification of PBLs, there is only a limited number of studies that have investigated the relationship between PBLs and BC risk factors. Nevertheless, it remains unclear as to which lesions should be completely resected and which should be followed up (1, 2, 3, 4). In this study, we aimed to investigate the relationship between the final pathology outcomes of PBLs and other risk factors of BC.

Materials and Methods

Patients who visited the surgical clinic of the university during the past 6 years and who with presented PBLs with or without atypia by fine/core needle aspiration biopsy were included in this study. Patients aged <18 years, whose file information could not be reached, and those without follow-up information were excluded from the study. Fibrocystic disease, fibroadenoma, normal breast tissue, and inflammation were classified as benign, and all cancer types were classified as malignant. We assessed the relationship between PBLs and BC risk factors such as the age, mass size, Body Mass Index (BMI), smoking habit, sports activity, BC family history, use of hormone replacement therapy, the number of pregnancies, and the duration of breastfeeding.

Statistical analysis

The sample size was calculated with Power (least) %80 and Type-1 error 0.05 for all variables. The Kolmogorov-Smirnov (n>50) and Skewness-Kurtosis tests were applied to examine whether the measurements in the study were normally distributed. Accordingly, parametric tests were applied since the measurements were normally distributed. In this study, descriptive statistics for continuous variables were expressed as the mean, standard deviation, and the minimum and maximum values. Categorical variables were described as number (n) and percentage (%). Independent t-test and one-way analysis of variance (ANOVA) tests were performed to compare the group mean values in continuous variables. Following the ANOVA, the Duncan post-hoc test was used to determine the different groups. Pearson’s correlation coefficients were calculated to determine the relationship among the variables. The chi-square test was employed to determine the relationship between the groups and among the categorical variables. Statistical significance level was considered as 5% in the calculations, and SPSS (IBM SPSS for Windows, ver.23) statistical package program was used for the calculations.

Results

The medical files of 77 cases were retrospectively reviewed. The descriptive properties are shown in Table 1. The median age of the patients was 38 (range: 19–74) years. The cut-off value of age was 35.5 years. The mean age of the PBL-WA patients was 40.98±12.74 years and that of PBL-WOA patients was 30.75±12.36 years (p=0.005). The mean age as per the final pathology was 33.66±10.17 years for the benign group and 50.33±12.15 years for the malignant group (p<0.001). The mean age as per the radiology features was 35.12±11.6 years for the benign group and 46.19±13.46 years for the malignant group (p<0.001). The mean age of the PBL-WA patients in the benign final pathology group was 34.89±8.84, while it was 51.04±11.90 years for the PBL-WA patients in the malignant final pathology group. In both the groups, the mean age was greater in the malignant group than in the benign groups (Table 2).

The mean size of the mass was 2.53±1.33 (1–6) cm, and the cut-off value of the mass size was 2.5 cm. The mean mass size for the PBL-WA patients was 2.64±1.37 cm, while it was 2.13±1.15 cm for the PBL-WOA patients (p=0.171). The mass size as per the final pathology was 2.25±1.22 cm in the benign group and 3.17±1.37 cm in the malignant group (p=0.004). The mean mass size was greater of the malignant lesions as per the fine/core needle aspiration biopsy, final pathology, and radiological imaging. The mean BMI value was 26.8±4.4 kg/m² (range: 18.8–35.1), and the cut-off value was 25.4. The corresponding value was 27.6±4.2 kg/m² for the PBL-WA patients and 23.8±3.9 kg/m² for the PBL-WOA patients (p=0.002). BMI as per the final pathology was 25.1±3.8 kg/m² in the benign group and 30.6±3.0 kg/m² in the malignant group (p=0.001). The mean number of children was 3.08±2.1 (0-8) in the PBL-WA group and 2.1±2.4 (0–7) in the PBL-WOA group (p=0.156). The mean overall total duration of breastfeeding was 51.8±41.7 months (0–156), and it was 56.9±41.5 months in the PBL-WA group and 32.5±37.4 months in the PBL-WOA groups (p=0.036). The cause of the lower number of children in the benign group was the lower patient age (Table 2).

Smoking habit was reported in 66.2% (n=51) of the patients. A total of 48 (94.1%) patients were included in the PBL-WA group and 3 (5.9%) patients in the PBL-WOA group (p=0.001). In the PBL-WA group, 68.9% (n=42) of the patients had no history of sports activities, 16.4% (n=10) had a history of sports activities once a week, and 14.8% (n=9) had a history of sports activities twice a week. In the PBL-WOA patients, 43.8% (n=7) of the patients had no history of indulging in sports activities, 37.5% (n=6) of the patients had a history of indulging in sports activities once a week, and 18.8% (n=3) of the patients had a history of indulging in sports activities twice a week (p=0.129). In addition, 83.1% (n=64) of the patients had no BC family history, while 16.9% (n=13) had a BC family history. Moreover, as per the final pathology, there were four (30.7%) patients in the benign group and nine (69.3%) patients in the malignant group (p=0.001). In addition, 88.3% (n=68) of the patients did not use hormone replacement therapy (HRT), while 11.7% (n=9) did (Table 3).

The malignancy rate of the PBL-WA patients was 37.7% (n=23), while it was 6.3% (n=1) in the PBL-WOA patients as per the final pathology (p=0.016). Breast-conserving surgery or mastectomy and sentinel lymph node dissection was performed in 19 (79.1%) patients, axillar lymph node dissection in five (20.9%)patients, and modified radical mastectomy in five (20.9%) patients. The positive predictive value for malignant lesions in the PBLs was 90.2%, negative predictive value was 73%, and accuracy was 84.4% for radiology (p=0.001). Twelve patients (15.6%) did not undergo surgery, and the follow-up time was 4.72±2.49 years. Six of these patients (50%) had PBL-WOA patients and the other six (50%) were PBL-WA patients. The mean age of the patients was 34.75±10.29 years, and the mean size was 1.67±0.78 cm. Both the mean age and size were lower than the cut-off value. None of them were diagnosed with malignancy during the follow-up time.

Discussion and Conclusion

Benign breast lesion can be classified as non-PBLs, PBL-WOA, and PBL-WA. These lesions were detected more frequently because of the widespread use of mammography, which makes it is important to identify patients at risk for BC. PBLs, especially containing atypia, are the risk factors for both non-invasive and invasive BC. In the PBL-WOA patients (e.g., complex fibroadenoma, moderate or floride hyperplasia, sclerosing adenosis, and intraductal papilloma), there is a slight increased risk for BC [relative risk (RR): 1.3–2]. The risk is greater in PBL-WA patients (such as atypical lobular hyperplasia and atypical ductal hyperplasia; RR: 4–6). When the atypia is multifocal, the risk increases by 10 times (4, 5, 6). In our study, the rate of malignancy of PBL-WA patients was greater than that of PBL-WOA patients as per the final pathology.

While the relationship of PBL-WOA and BC does not change with age, it is stronger in postmenopausal patients (6, 7). In our study, however, we observed a significant effect of age on the type of PBLs. The mean age of the PBL-WA patients was 40.98±12.74 years and that of the PBL-WOA patients was 30.75±12.36 years (p=0.005). As per the final pathology, the mean age was 33.66±10.17 years in the benign group and 50.33±12.15 years in the malignant group (p<0.001). As the risk of BC increases with age, age was noted as an important factor in PBLs. Malignant lesions were recorded in the advanced age in both the groups (patients with PBL-WA and patients with malignant pathology result) (Figure 1).

Renshaw et al. (8) reported no correlation between the size of lesion and atypical ductal hyperplasia or ductal carcinoma in situ. However, the size of lesions diagnosed as carcinoma was significantly greater than that of lesions diagnosed as PBL-WA (p<0.001). In our study, the mean pathological tumor size was 2.25±1.22 cm in the benign group and 3.17±1.37 cm in the malignant group (p=0.004). The size of the mass was larger in all malignant patients (Figure 2).

Several past epidemiological studies have shown that being overweight and/or obese, indicated by BMI in postmenopausal women, is a risk factor for BC development (9, 10, 11). BC is more common in obese women (BMI >30 kg/m²) (12). When postmenopausal women lose ≥10 kg, they are at a lesser risk than those who do not lose weight (7, 13). In our study, while the BMI was 27.6±4.2 kg/m² (n=61) for the PBL-WA patients, it was 23.8±3.9 kg/m² (n=16) for the PBL-WOA patients (p=0.002). In addition, as per the final pathology, BMI was 25.1±3.8 kg/m² (n=53) for the benign group and 30.6±3.0 kg/m² (n=24) for the malignant group (p=0.001) (Figure 3). Several studies have also shown that pregnancy and breastfeeding have a protective effect against BC (14). In our study, the mean number of children and the total duration of breastfeeding were lower in the benign groups due to the young age of the patient. The relationship between smoking and BC is contradictory. Although very different results have been reported in the literature, it is believed to increase the risk associated with some other factors (15, 16). Positive smoking history was 78.7% (n=48) for the PBL-WA patients and 18.8% (n=3) for the PBL-WOA patients (p=0.001) (Figure 4).

Increased physical activity, especially in premenopausal women, is associated with a reduced risk of BC (7). Lynch et al. (17) indicated an average of 25% reduction in BC risk among physically active women when compared with the least active women in a meta-analysis of 73 studies on the relationship between physical activity and BC. In our study, the percentage of patients with no sport activities was more in the malignant group than in the PBL-WA group as per the final pathology.

Patients with a family history showed a higher risk of developing BC, but the effect of PBLs with a family history has been discussed in the literature. The possibility of developing age-related BC in 10 years in women with a family history and proliferative breast disease is one in 2000 at the age of 20 years, one in 256 at 30, one in 67 at 40, one in 39 at 50, and one in 29 at 60 (7, 18, 19). A family history of maternal BC has not been found to be related to the degree of atypia or fibrocystic breast disease in most hospital-population-based studies (20, 21, 22). The family history of BC has very little effect on the risk of developing BC in patients with non-PBLs; however, there is an 11-fold increased risk in patients with PBLs presenting with atypia (23). In our study, the percentage of patients with BC family history was greater in the malignant group than in the PBL-WA group, as per the final pathology.

Both the World Health Organization and the One Million Women Study have revealed that women who received HRT had an increased risk of developing BC. however, as per epidemiological studies, no relationship has been established between the use of HRT and the risk of developing BC. Although a relative increase in risk of 1.24 was reported by a few large-scale studies, this relationship has not been revealed in the two recent studies (24, 25, 26, 27). In our study, no statistically significant risk was noted between the use of HRT and the development of PBL-WA.

The limitation of the present study is that it was a single-center study with a smaller sample size.

In conclusion, our results indicate that age is an extremely important aspect in assessing PBLs. The patient age was statistically significantly greater in those with malignant lesions in all groups, such as the radiological imaging features of the lesions, fine/core needle aspiration biopsy results, and the final pathology. The lesion size, BMI, smoking habit, and family history of BC were also more frequent in the malignant group. The rate of sports activity was lower in the malignant groups. The number of pregnancies and the total breastfeeding time were smaller and lower, respectively, in the benign groups, possibly due to the lower average age of the patients. The use of HRT showed no effect on the benign and malignant lesions. Thus, it seems necessary to evaluate patients individually when evaluating PBLs. It is therefore recommended to evaluate PBLs together with BC risk factors.

Ethics Committee Approval: Ethical approval for this study was obtained with regard to the Ethical Principles for Medical Research Involving Human Subjects (the Helsinki Declaration) from the Local Ethics Committee of Van Yüzüncü Yıl University, Turkey with the registration number of 2019/05-09.

Informed Consent: Written informed consent was not recieved due to the retrospective nature of the study.

Peer-review: Externally-peer reviewed.

Authorship Contributions

Concept: O.T.; Design: O.T., R.E.; Supervision: O.T., S.E., Ü.H.İ.; Resources: A.R.K., A.Ö., F.A., S.B, İ.Ö., E.Ş.; Materials: S.B., İ.Ö., E.Ş.; Data Collection and/or Processing: O.T., S.B., Ü.H.İ., A.R.K., F.A.; Analysis and/or Interpretation: O.T., S.E., Ü.H.İ., A.Ö.; Writing Manuscript: O.T., R.E., S.E.; Critical Review: O.T., S.E.

Conflict of Interest: The authors declare no potential conflict of interests with respect to the research, authorship, and/or publication of this article.

Financial Disclosure: The authors received no financial support for the research, authorship, and/or publication of this article.