BC has surpassed lung cancer as the most commonly diagnosed type of cancer among women, with ~2.26 million new cases and ~685,000 deaths recorded globally in 2020. These numbers represent 24.5% of all new cancer cases and 15.5% of cancer-attributed mortality in women, respectively (1). In the US, the new cases of BC have also markedly increased, reaching top levels among all new cancer cases among women, with ~282,000 cases estimated in 2021 (30%); BC also represents the 2nd leading cause of cancer-related mortality, with 43,600 deaths (15%), coming only second following lung cancer (2).

The BC incidence rates have consistently increased during the decades between 1970-2000 in a number of industrialized countries, likely reflecting changes in lifestyle associated with civilization and increased detection via mammographic screening. In the 2000s, the incidence rates had reached a plateau with slight decreases in some countries (3), which were largely attributed to a reduction in the use of menopausal hormone therapy and also possibly to a stabilized screening participation rate (4-6). However, in the US, the BC incidence rates have continued to increase by ~0.5% annually in recent years (7). This is attributed at least in part to increased body weight (8).

The prevalence of obesity, which is defined as a body mass index (BMI) ≥30, has markedly increased in recent decades; globally, the incidence of obesity has almost tripled since the 1970s (9). Worldwide, >1.9 billion adults aged ≥18 years were overweight, accounting for 39% of the population in 2016. Of these, >650 million, 13% of the population, were obese (9). In the US, according to the National Health and Nutrition Examination Survey (NHANES) data from 1999-2000 through 2017-2018, the age-adjusted prevalence of obesity among adults aged >20 years increased from 30.5 to 42.4%, representing a ~200% increase from 1976-1980. The prevalence of severe obesity (BMI ≥40) increased from 4.7 to 9.2%, and it was extremely rare before the early 1970s (10,11). Although the increasing trend in the prevalence of obesity has recently exhibited a decrease, a further increase, up to ~50% by the year 2030, is still projected (12,13).

Obesity is associated with a variety of health risks, and severe obesity further increases the risk of obesity-related complications (14-16). Obesity is a risk factor for type 2 diabetes (17-20), cardiovascular diseases (21,22), inflammatory bowel diseases (23,24) and overall mortality (25,26). According to the examination of the association between obesity and US adult mortality in recent decades (1986-2006), scientists have reported that obesity accounted for 18% of deaths among African Americans and Caucasians between the ages of 40 and 85, which is significantly greater than the ~5% previously considered (27). In terms of cancer, the International Agency for Research on Cancer (IARC) has identified 13 types of cancer associated with overweight and obesity: Meningioma, multiple myeloma, adenocarcinoma of the esophagus, cancers of the thyroid, post-menopausal breast cancer (post-BC), gallbladder, stomach, liver, pancreatic, kidney, ovarian, uterine, and colorectal cancer (28). In US, the Centers for Diseases Control and Prevention (CDC) have reported that overweight and obesity constitute 40% of the cancer cases diagnosed (29). Therefore, the rising rate in obesity is currently a serious public health concern.

Well-established risk factors for BC are numerous, including non-modifiable factors such as age, race, family history, etc., and modifiable factors such as diet, physical activity, body weight status, alcohol consumption, hormonal status, parity and factors leading to greater birthweight or greater linear growth, etc. (30). Of these modifiable risk factors, the body weight status is of high significance as regards its association with the incidence and prognosis of cancer (31), as well as the current obesity epidemic (9,11).

The associations between overweight/obesity and BC have become complex. The body weight status is significantly associated with the overall risk of developing BC (32). A higher BMI, higher overall energy intake and lower physical activity levels have been shown to be associated with an increased risk of developing post-BC in prospective cohort analyses (33). It was originally speculated that the body weight status was also positively associated with the risk of developing pre-BC (33). However, in recent studies, although overweight or obese weight status have been continuously shown to be positively associated with the risk of developing post-BC, this association has been indicated to be reversed in women of childbearing age (34-36). A meta-analysis of 34 cohort studies including >2.5 million women and almost 24,000 post-BC subjects reported that, accompanying each 5-unit increase (5 kg/m²) of BMI in overall adulthood, the relative risk of post-BC increased by 12% (37). For pre-BC, a meta-analysis performed by the World Cancer Research Fund/American Institute of Cancer Research (WCRF/AICR), with 37 studies and 16,371 cases, revealed a statistically significant 7% decreased risk per each 5-unit increase in BMI in overall adulthood (30). It is currently an accepted concept that a well-established positive association exists between obesity and post-BC, whereas a reverse association exists prior to menopause (38,39) (Fig. 1).

A significant challenge for understanding the cellular and molecular mechanisms of action by body fatness is the apparent increased risk of developing post-BC; however, body fatness seems to have a protective impact on pre-BC, and a high BMI in young adulthood seems to have a protective effect on post-BC. There is no single mechanism through which obesity mediates breast tumorigenesis, and different molecular mechanisms are expected in the development of pre- and post-BC. A number of studies have been conducted in order to understand the positive association between obesity and post-BC, and several mechanisms governing this association have been proposed and defined, including adipose tissue-driving circulating hormones and an obesity-associated chronic low-grade inflammatory state (12,40). However, the mechanisms responsible for the associations between obesity and pre-BC are less understood, and a recent study by the authors demonstrated a potential epigenetic mechanism: Obesity influences folate metabolism and leads to an elevation in breast tissue folate levels, preventing the development of pre-BC by altering global methylation (41). Further details on this matter are discussed below, as well as in the previous study (41).

In cancer, two contradictory changes in DNA methylation patterns have been characterized: Global hypomethylation and gene-specific hypermethylation (59). Global hypomethylation facilitates cancer development and progression through chromosomal instability (60) and the activation of oncogenes (61) or the silencing of tumor suppressors (62). The global hypomethylation level of peripheral blood leukocyte DNA has been suggested as a potential biomarker for the risk of developing BC (63). In a methylome mapping at the single nucleotide resolution of a low-passage BC cell line (HCC1954) and primary human mammary epithelial cells, a widespread DNA hypomethylation was reported in BC cells, primarily at partially methylated domains in normal breast cells (62). Genes [e.g., the DNA repair gene, O-6-methylguanine-DNA methyltransferase (MGMT)] within these regions were largely silenced in BC cells and the loss of DNA methylation in these regions was accompanied by the formation of repressive chromatin, indicating a widespread DNA hypomethylation for gene regulation through chromatin alteration (62). A recent systematic analysis found an association between global blood-derived DNA hypomethylation with a higher risk of developing BC, and the strength of the association tends to be weak for long period follow-up, indicating that global DNA methylation may be a short-term predictor of the risk of developing BC (64).

Compared with global hypomethylation, gene-specific hypermethylation and the silencing of tumor suppressor genes in cancer have received increasing attention over the past decades. A list of genes hypermethylated in BC has been identified and methylation signatures have been established with whole epigenome approaches (65,66). A number of tumor suppressors are hypermethylated in BC cells and primary mammary tumors (43). Their biological function includes DNA repair, cell-cycle regulation, proliferation and apoptosis, cellular homeostasis, cell adhesion and invasion, etc. A list of commonly reported genes with differential methylation in BC is presented in Table I (66-69). A known example is the BRCA1 gene, which is normally expressed in cells of the breast where, together with the BRCA2 gene, are responsible for repairing damaged DNA (70,71). When BRCA1 is hypermethylated, it gives rise to the same pattern of gene expression as the one of BRCA1-mutated BC, a common type of hereditary BC (72). As aberrant promoter methylation patterns in tumor suppressors are common phenomena in BC, several studies have already evaluated the applicability of gene-specific hypermethylation as a predictive (73), diagnostic (74-76) or prognostic (77-79) biomarker in BC. For instance, a whole-genome methylation capture sequencing analysis of small amounts of DNA isolated from formalin-fixed, paraffin-embedded tissue from triple-negative BC (TNBC) and matched normal samples identified TNBCs into three distinct methylation clusters associated with a better or worse prognosis (79). These data highlight the prognostic potential of DNA methylation in TNBC, and provide valuable tools for the management of TNBC (79).

Diets with an enriched fat content are associated with an elevated body weight and studies have also shown that a diet high in saturated fat is associated with aberrant DNA methylation. The study by Perfilyev et al examined studied the impact of a 7-week intake of extra amounts (+750 kcal/day) of saturated fat on the genome-wide DNA methylation in the subcutaneous adipose tissue in young healthy humans. They demonstrated that the intervention did not differ in the degree of body weight gain, but rather in the degree of DNA methylation of 125 genes [e.g., adiponectin, C1Q and collagen domain containing (ADIPOQ)] differentially methylated within adipose tissue (83). Another study demonstrated that the short-term high-fat overfeeding of healthy young men induced DNA methylation alterations in 6,508 genes in skeletal muscle biopsies. Among the top 20 most significant genes were dynamin 2 (DNM2), MGMT, glucose transporter type 3 (GLUT3), mannose receptor C-type 1 (MRC1) and acetyl-CoA acetyltransferase 2 (ACAT2) (84). In contrast to a high fat diet, caloric restriction has bene shown to reduce the impact on DNA methylation. The study performed by Milagro et al (85) demonstrated that an 8-week caloric restriction altered DNA methylation at several sites of ATPase phospholipid transporting 10A (ATP10A) and WT1 in peripheral blood mononuclear cells. Moreover, in another study, a 6-month caloric restriction resulted in a >3% reduction of body fat and induced the hypermethylation of phospholipase C eta 2 (PLCH2) and PR/SET domain 8 (PRDM8) in subcutaneous adipose tissue biopsies (86).

Physical activity is directly associated with obesity. There is emerging evidence supporting that physical activity influences DNA methylation in humans (87). Acute high-intensity exercise has been demonstrated to elicit the hypomethylation and elevated expression of PPARG coactivator 1α (PGC-1α), pyruvate dehydrogenase kinase 4 (PDK4) and peroxisome proliferator-activated receptor δ (PPAR-δ) in skeletal muscle in both humans and mice (88). A genome-wide analysis of DNA methylation in the muscle of trained mice (5 days/week for 4 weeks) identified >2,700 genes with significant methylation changes in their putative promoter regions compared with sedentary controls (89). A lifelong physical activity study demonstrated that DNA methylation was significantly lower in >700 promoters of genes in the skeletal muscle of physically active than inactive men, and these genes were involved in metabolism, myogenesis, contractile properties and oxidative stress resistance (90). A systematic review of 25 studies concluded that both acute and chronic exercise significantly affected DNA methylation, in a highly tissue- and gene-specific manner; among genes whose methylation levels were found to be significantly altered after exercise were those involved in metabolism [e.g., PGC-1a and glutamate dehydrogenase 1 (GLUD1)], muscle growth [e.g., myocyte enhancer factor 2A (MEF2A)] and inflammation [e.g., PYD and CARD domain containing (PYCARD, or ASC)] (87).

Although not differentiated by the menopausal status, several studies have investigated the associations between BMI and the risk of developing BC as regards gene-specific DNA methylation levels. For instance, a number of studies have demonstrated that the methylation status of CpG islands in the promoter regions of the BRCA1 gene is aberrant in patients with sporadic breast tumors when compared with healthy females or patients with benign diseases (100-106). A meta-analysis of >40 studies demonstrated that the frequency of BRCA1 promoter methylation was significantly higher in BC patients than in healthy controls, and this BRCA1 methylation was also associated with diminished levels of BRCA1 protein expression, metastasis, histological grade 3 and the triple-negative phenotype (71). By contrast, in a healthy cohort of dominantly obese female nurses aged 40-60 years, obesity was not associated with BRCA1 or BRCA2 promoter hypermethylation (107). These disparate results indirectly that indicate obesity plays a mediating role in the association between gene-specific methylation and BC. In another study, BC-specific mortality was found to be higher in obese women with promoter methylation in the APC gene [hazard ratio (HR), 2.47; 95% confidence interval (CI), 1.43-4.27] (108). However, the data are too limited for a conclusion to be drawn regarding the mechanisms through which obesity mediates gene-specific DNA methylation in terms of the development of BC, particularly as regards the differential impact on pre-vs. post-BC.

Similar to gene-specific DNA methylation, the understanding of the mechanisms through which obesity mediates global DNA methylation are also limited. Variations in global methylation measurements (63,109-111) further complicate the determination of the effects of obesity on global DNA methylation and BC. High performance liquid chromatography-based (HPLC) detection is considered the ‘gold standard’ of global DNA methylation (112); however, due to the considerable cost and need for specific laboratory conditions, the use of HPLC for DNA methylation profiling is less common. Two of the most popular substitutions for measuring global DNA methylation are the methylation status of long interspersed nuclear element-1 (LINE-1) and the luminometric methylation assay (LUMA). As more than one-third of methylation within the human genome is found in the CpG-rich sequences of LINEs, these transposable elements are considered a valid surrogate marker for global methylation (113,114). LUMA measures DNA methylation in CCGG sequences and provides a robust estimation of the overall 5-methylcytosine (5-mC) content in dinucleotide CpG sites in the whole genome (115,116). Of these two methods (LUMA and LINE-1), only LINE-1 assay data are well-associated with HPLC-derived measurements, and LINE-1 is often recommended over the use of LUMA alone to assess whole genome methylation patterns (117).

It has been indicated that global DNA methylation modifies the association between obesity and BC. A study on >1,300 patients with ~15 years of follow-up, demonstrated an increased all-cause mortality (HR, 1.81; 95% CI, 1.19-2.74) and BC-specific mortality (HR, 2.61; 95% CI, 1.45-4.69) among obese patients with the lowest LUMA levels (global hypomethylation) (108). Although not directly linking LINE-1 with BC, a previous study on 156 individuals found that LINE-1 methylation was positively associated with a healthier lifestyle, yet inversely related to body fat mass in healthy young individuals (118). However, this result is inconsistent with the findings of other studies, which indicated that an elevated BMI was associated with lower LINE-1 methylation (119,120) or was not associated with LINE-1 (121). Another cross-sectional study on 289 healthy postmenopausal women who participated in the Alberta Physical Activity and Breast Cancer Prevention trial demonstrated that, after adjusting for important confounders, LINE-1 methylation was positively associated with BMI (P=0.03) and multiple other adiposity parameters, including intra-abdominal fat area, body fat percentage, fat mass, waist circumference, hip circumference, current weight status, body weight at age 20 and adulthood weight gain (122).