Clinicopathological correlation between trophic-adipose levels and poor prognosis outcomes in Brazilian women diagnosed with breast cancer

Breast cancer is the most common malignant neoplasm in women.1 Determinant risk factors include reproductive history, family history, and lifestyle. These factors are responsible for clinical and pathological differences.2 Although widely studied, the main risk factors associated with the occurrence of breast cancer in women from different regions of Brazil are poorly understood perhaps because of the large geographical area involved in Brazilian studies.

Previous Brazilian studies suggest classic risk factors for breast cancer development and aggressiveness such as aging and menopausal status.3 4 Other studies show more complex associations such as more aggressive disease and worse prognostic tumors such as triple-negative tumors in overweight and/or obese women.5 Factors such as social vulnerability6 and history of psychological stress7 are also possible risks particularly in southern Brazil. These factors do not explain the occurrence of disease alone and have been widely reported as associated with each other.

Thus, we seek more detailed information about the profile of breast cancer patients in Brazil where studies on regional risk factors are still scarce and inconclusive. There is no official documentation of such data from specific regions of the country including southwestern Paraná. Thus, this study characterized the epidemiological profile and possible regional risk factors identified in women diagnosed with breast cancer treated between 2015 and 2017 in a cancer referral hospital of 27 municipalities that make up the 8^th Regional Health of Paraná.

This is a retrospective descriptive observational study which proposal was submitted to the Institutional Ethics and Human Research Committee approved under the CAAE (certificate of presentation of ethical appreciation) number 35524814.4.0000.0107 and under opinion No. 810.501. All participants gave informed consent on the study objectives. Their anonymity was ensured, and they could withdraw at any time. The inclusion criteria were patients referred for surgery with lesions suggestive of unilateral infiltrative ductal carcinoma (ICD) at any clinical stage attended by the Francisco Beltrão Cancer Hospital from May 2015 to August 2017. These patients were from the 8^th Regional Health of Paraná covering an estimated population of 350,000 inhabitants located at 27 municipalities (Figure 1).

Patients who do not meet this criterion were excluded. Thus, from this initial cohort of 200 women, there were 126 women with a histologicallyconfirmed diagnosis of breast cancer by biopsy. These women had complete clinicopathological data for subsequent frequency analysis. The medical records were consulted for data collection.

The data were compared for possible existing correlations with age at diagnosis; tumor size; histological grade; expression pattern of receptors and molecular subtypes; lymph node invasion; presence of angiolymphatic emboli; TNM (tumor, lymph nodes, and metastasis) clinicopathological staging; menopausal status; body mass index (BMI); and recurrence risk stratification. Data were categorized and analyzed using Statistical Package for Social Science (SPSS) statistical software (version 25.0.0, IBM) to obtain the frequencies and apply the chi-square test and logistic regression analysis. Only the significant correlations/associations were shown in the results, considering p<0.05 as significant.

This study compiled sequential data from 200 serially collected biopsy specimens from women presenting with lesions suggestive of breast cancer diagnosed by imaging exams such as mammography, ultrasound, or magnetic resonance imaging (MRI) as well as physical examination. Ten patients were excluded due to a lack of clinicopathological data. Of the 190 samples, 127 were confirmed as breast cancer (66.8%). One patient was excluded for a lack of sufficient data leaving 126 participants.

Since there was no statistically significant difference between the overweight and obese groups (the obese group was only 6% of the sample), we decided to combine these two BMI categories into one group and compare them with the eutrophic patients. Table 1 shows the main clinicopathological findings regarding such groups.

We found that 57.1% of the patients were overweight at diagnosis, and the mean BMI was 27.54kg/m² (18.22kg/ m² to 44.15kg/m², Figure 2). Regarding the frequency of molecular subtypes of tumors, 68.3% of patients with triple-negative tumors fell into the overweight subgroup with an equal distribution of the other subtypes between the eutrophic and overweight groups. Further, overweighed/obese patients had larger tumors: 62.5% of tumors with diameters between 2 to 5cm and 94.1% of tumors with diameter greater than 5cm.

There was also a predominance of high histological grade tumors in overweight/obese patients (70.4% of the tumors diagnosed in the study within this category), with a high recurrence rate (65%). Statistical analyses (Table 2) showed significant associations only in the overweight/obese patients, including the presence of intermediate histological grade luminal B subtype tumors (β=0.630; 95%CI: 0.184-1.075 and p=0.006). There was a positive association between tumors of intermediate grade and tumors between 2 and 5cm in diameter (β=0.294; 95%CI=0.057-0.531 and p=0.016). There was positive association between the presence of angiolymphatic emboli and high stratification. There was a higher-risk of recurrence in this group (β=0.169; 95%CI: 0.039-0.298 and p=0.012). No significant association was found in the eutrophic group.

In recent decades, obesity has emerged as an important risk factor associated with the development of several cancers including breast cancer. In postmenopausal women, excess body fat is directly implicated in the development of triple negative tumors with worse prognostic outcomes regardless of age.8 A wide range of mediators are implicated in this context including molecules that perpetuate chronic inflammation that have a wellestablished role in breast cancer carcinogenesis and progression.9 10

Similar to other studies, we found that Brazilian women with a BMI over 30kg/m² have an elevated risk of developing breast cancer.3 In addition, both obesity and overweight are risk factors for the development of triple negative tumors in a Brazilian population.5 In fact, we found that the excessive body weight seems to be the main risk factor in this population. Adipose tissue is known to fuel a chronic inflammatory state that can influence breast tissue transformation via the production of carcinogenesis-promoting mediators resulting in worse outcomes.11

A predominance of triple negative tumors in obese/ overweight women was identified in our study. The presence of triple negative tumors in women with excess body fat is already well established in the literature and is common in people of African descendants.12 We note that the population studied here is predominantly Caucasian (over 90%), which rules out the association between a high prevalence of triple negative tumors and race or BRCA (breast cancer) gene mutations.13 Our patients were screened for germinative pathological variants and we found a low prevalence of pathogenic variants in BRCA (2 cases).

Excessive body weight was also associated with tumors larger than 2cm. This finding suggests that deregulation of fat metabolism may occur locally in breast tissue and lead to carcinogenesis. Indeed, breast tissue is in a continuous proinflammatory state in high BMI subjects - especially in patients with excess visceral fat.14 Furthermore, localized excess breast fat can induce adipocyte death and activate macrophage recruitment including activation of important pathways that maintain chronic inflammation such as NFĸB (factor nuclear Kappa B).15 Such metabolic and immunological changes would eventually generate a microenvironment that is conducive not only to carcinogenesis but also cellular phenotypic transformation, increased proliferation, and increased tumor mass.16

Malignant transformation also seems to be associated with excess fat, and this could explain the high prevalence of high histological grade tumors found in the overweight/obese cohort. Mediators such as leptin - whose production is increased proportionally to the increase in body fat - are positively associated with the development of highgrade breast tumors17 as well as the occurrence of triple-negative tumors18 seen here. In addition, increased waist circumference in women with breast cancer has also been described as a predictor of the development of high-grade tumors19 suggesting that fat may affect the process of breast tissue differentiation. In this sense, experimental evidence suggests that suppression of endogenous lipogenesis may reverse the malignant breast cancer cell phenotype and reprogram breast cells to follow the normal process of cell differentiation.20

Our study showed some important associations between the parameters evaluated in overweight women. There was a positive association between tumors of intermediate histological grade and luminal molecular subtype B. There was also correlation of intermediate grade with 2-5cm tumors. Both associations suggest that tumors formed in the presence of excess body fat have greater proliferative capacity. This implies the formation of larger masses and accelerated cellular de-differentiation.

There was also an association between the presence of angiolymphatic emboli and the high-risk of recurrence in the overweight/obese cohort. Increased embolus formation is common in both cancer and obesity alone due to the endothelial activation triggered by chronic inflammation.21 Such formation may be correlated to the development of hypoxia in the tumor tissue - this process that can be potentially aggravated in overweight or obese patients with breast cancer and hypercoagulability states.22

Despite the total number of individuals included in the study is good, the small sample size observed for each group is an important limitation for further conclusions. Also, the retrospective design limited data collection limited for some clinical parameters as survival information and chemotherapy response.

Immunological and endocrine changes in the tumor microenvironment due to excess body fat might trigger the development of more proliferative tumors, larger tumors, and tumor with accelerated cell de-differentiation. The endothelial injury caused by a continuous and systemic proinflammatory state due to obesity can lead to neoplastic cell dissemination leading in turn to metastatic disease and more advanced clinical staging.