1. Introduction

Endometrial cancer (EC) is an adenocarcinoma that originates from the epithelial cells lining the uterine cavity. The tumor microenvironment surrounding these cells comprises stromal cells, endothelial cells [1] and many different types of immune cells [2], all of which can influence cancer progression and response to treatment. Although most ECs are early stage and confined to the uterus, others spread by invading the myometrium and metastasizing to distant sites such as lymph nodes, liver and lung [3]. According to GLOBOCAN 2018 statistics, EC is the sixth most common cancer and the 11th leading cause of cancer death in women worldwide, with 382,069 new cases and 89,929 deaths [4].

2. Classification of Endometrial Cancers

EC has been classically categorized into two clinicopathological subtypes; Type I and Type II. Type II tumors are generally more invasive, estrogen and progesterone receptor (ER/PR) negative and confer a poor prognosis; but account for less than 15% of all cases ^[5][6][5,6]. Type II ECs include grade 3 endometrioid, papillary serous, clear cell, and carcinosarcoma histologies [7]. In contrast, most EC cases are Type I tumors that are frequently low grade endometrioid tumors, confined to the uterus, ER/PR positive, and have higher survival rates following treatment with primary surgery [6]. Since most EC cases are Type I, the majority of studies sited in this review refer to or contain data from Type I EC, unless otherwise stated. However, recent genetic investigations have supported re-classification of EC into 4 molecular subgroups; (1) DNA-polymerase epsilon (POLE) (ultramutated), (2) microsatellite instability hypermutated, (3) copy-number low (microsatellite stable), and (4) copy-number high (serous-like) [8]. Each subgroup has prognostic significance, with the POLE group having the best prognosis (>95% progression-free survival) and the copy-number high group the worst (5-year progression-free survival of 50%). These molecular subgroups are associated with body mass index (BMI), with women in the POLE cluster having the lowest BMI and those in copy-number low cluster having the highest BMI, suggesting that obesity may impact the genetic landscape of endometrial tumors [9].

3. Risk Factors for Endometrial Cancer

Recent epidemiological studies have shown an increasing incidence of EC, especially in countries with rapid socioeconomic transitions [10]. The rate of new cases of EC is expected to rise due to an aging population and an increased prevalence of risk factors, particularly obesity [11]. In addition to obesity, a number of factors have been attributed to an increased likelihood of developing EC, including but not limited to; advancing age, late onset menopause, lower age of menarche, chronic anovulation including polycystic ovarian syndrome, estrogen therapy in the absence of progesterone, tamoxifen therapy, hereditary predisposition (Lynch syndrome), and nulliparity ^{[12][13][14][15]}[12–15]. Several of these factors influence the length of time and level of exposure the uterus has to estrogen and progesterone.

A number of Mendelian Randomization (MR) studies have identified causal factors for EC (as reviewed ^[16][17][16,17]). MR is an analytical method that uses genetic determinants (variants), typically single nucleotide polymorphisms (SNPs), as instrumental variables for a modifiable risk factor. The MR approach is considered to be less affected by confounding factors or reverse causation but is dependent on assumptions [17]. One study showed that a genetically-predicted increase in age of menarche, adjusted for genetically-predicted body-mass index (BMI), was associated with a lower risk of EC [18]. Another showed that variants in CYP19A1 (the gene that encodes aromatase/estrogen synthetase) were associated with increasing estradiol levels in post-menopausal women, and risk of EC, in women of European ancestry [19]. SNPs associated with obesity (BMI), but not waist:hip ratio, were also shown to be associated with EC, indicating that obesity is a causal factor for EC [20]. Genetically-predicted higher fasting insulin levels (using 18 SNP variants) and post-challenge insulin levels (using 17 SNP variants), but not fasting glucose (using 36 SNP variants) or Type 2 diabetes (using 49 SNP variants), were associated with increased risk of EC [21] (Table 1). A more recent study by O’Mara et al. included the most numbers of cases and controls to date; 12,906 endometrial cancer cases and 108,979 country-matched controls of European ancestry [22]. This study confirmed previous findings (higher BMI associated with increased EC risk and later menarche with lower EC risk) and demonstrated that the protective effect of later menarche is partially mediated by the known relationship between lower BMI and this factor [22]. Overall, these genome-wide association studies may provide vital information to those proposing the development of a risk prediction scoring system for women at high risk of EC [23]. A scoring system such as this could enable prophylactic treatment to reduce the incidence of EC, particularly those with Type I EC [23].

Table 1. Hyperglycemia and Endometrial Cancer.

* overlapping populations. § Diabetics and patients with blood glucose > 125 mg/dL (~6.9 mmol/L) were excluded from study.

3.1. Links between Obesity and Endometrial Cancer

Worldwide, the prevalence of obesity [body mass index (BMI) > 30 kg/m²] in women has increased five-fold in the last four decades [40]. In women, it is estimated ^[40]that 20% of all cancer-related deaths are due to obesity, and of these, EC is the most strongly associated [41,42]. EC has the strongest association with obesity of all malignancies with a population attributable fraction (PAF) of 42.4% in the Oceania population (including Australia and New Zealand) and 56.8% in the US population [43,44]. Obese women are 2–3 times more likely to be diagnosed with EC [45] and the age of diagnosis of EC is inversely correlated with BMI [46]. Each 5 kg/m² increase in BMI correlates to a large increase in EC risk, with most observational studies reporting a 200%–400% increased risk of developing EC in individuals with BMI ≥25 kg/m² [47]. Calle et al. also reported a 6.25-fold increased risk of uterine cancer-related death for morbidly obese women compared to those within normal range of BMI [41]. Bariatric surgery is an effective treatment for weight loss for morbidly obese patients and a scoping review by Aubrey et al. found EC risk reduction in obese women who underwent bariatric surgery [48]. 

Mechanisms linking obesity and cancer have been described in the literature ^[41][42][49,50]. Several of these have been proposed to link obesity to EC development and progression, including: (1) excess estrogen through aromatization of androstenedione to estradiol by adipose-derived aromatase ^[43][51], (2) altered secretion of adipokines by adipocytes, specifically lower levels of adiponectin and higher levels of leptin ^[44] [52] and visfatin ^[45][46][53,54], (3) insulin resistance with associated hyperinsulinemia, increased insulin-like growth factor 1 (IGF-1) and decreased IGF binding protein 1 (IGFBP-1) and sex hormone binding globulin (SHBG) [44], and (4) chronic low grade inflammation from increased levels of proinflammatory cytokines ^[47][48][55,56]. These mechanisms linking obesity and endometrial carcinogenesis are described elsewhere ^[49][50][57,58]. The role of obesity, as a component of metabolic syndrome in EC, is also described in-depth in another review ^[51][59].

3.2. Links between Hyperglycemia and Endometrial Cancer

Disorders associated with hyperglycemia (Type I and II diabetes mellitus) have an increased risk of EC, indicating that poor control of blood glucose may be an important contributor to the growth of these tumors in women. Three separate meta-analyses on this topic have demonstrated that diabetes mellitus is significantly associated with a twofold risk of developing EC ^[52][53][54] [60–62] and several epidemiology studies have also demonstrated that this association is independent of obesity ^[55][56] [60,63–65] ^[57](Table 1). A case-control study involving 942 cases and 1721 controls conducted by Zhang et al. demonstrated a twofold increase in EC risk in women with type II diabetes mellitus (T2DM) compared to their non-diabetic counterparts ^[58] [66]. Furthermore, hyperglycemia has been associated with EC independent of obesity ^[59][67]. In a previous study by Modesitt et al. comparing women with comparable morbid obesity levels with and without Type I EC, circulating glucose levels were higher in women with cancer (119.5 vs. 90.7 mg/dl for non-cancer; p = 0.049) [32] (Table 1). Interestingly, other serological factors, including estrogen and insulin, were not significantly different between the two groups [32]. Several large prospective cohort and case-control studies have also found an increased risk of EC with higher blood glucose levels [24,25,27–31,34,35] (summarized in Table 1), although the strength of these associations varies according to BMI, age, and menopausal status in some populations [24,25,27] (Table 1). One observational study did not find an association, however diabetic patients and patients with a fasting blood glucose ≥6.9 mmol/L at baseline were excluded [26] (Table 1). Observational studies are susceptible to confounding and as such, it is possible that hyperinsulinemia, rather than hyperglycemia, is responsible for the association between T2DM and EC risk, as supported by an MR study [21] (Table 1).

Glycosylated hemoglobin (HbA1c) is used as an indicator of blood glucose levels over the preceding 3 months ^[60][68]. In Australia, levels ≥6.5% are considered elevated ^[61][69]. While HbA1c is a more helpful indicator of long-term glycemic control than fasting or random blood glucose levels, few studies have examined the association between elevated HbA1c and EC (Table 1). However, two small case-control studies showed higher mean HbA1c in EC cases versus controls [36,39]; and a prospective cohort study in a predominantly Maori population (the Indigenous people of New Zealand) found a four-to-five-fold increase in EC risk with elevated HbA1c [38](Table 1). Overall, there is evidence to suggest that the chronic elevation of blood glucose may increase the risk of EC.