Diabetes and the social, biologic, and behavioral determinants of endometrial cancer in the United States

Endometrial cancer is a type of cancer that develops in the tissue lining of a woman’s uterus [1], with associated symptoms of pelvic pain, and excessive, prolonged, or irregular bleeding between menstrual periods [2]. Endometrial cancer is currently the 7th most common cancer and the 14th leading cause of death among women in the world [3] and in the United States (US) with 63,000 newly reported cases in the year 2022 [4, 5].

Diabetes, a chronic disease characterized by high levels of glucose in the bloodstream [6], has been reported to be associated with endometrial cancer [7‐15]. In this respect, a systematic review that assessed the relationship between diabetes mellitus (DM) and endometrial cancer, analyzed 29 eligible cohort studies, and reported a summary relative risk of 1.89 for women with DM compared to women without DM [15]. In addition, this study has also reported a summary incidence rate ratio of 1.61 for women with versus without DM [15]. Another research article reported results from a systematic review and meta-analysis conducted on 22 cohort and case-control studies identified 14 studies that detected significant association between DM and endometrial cancer. This article showed that diabetes was significantly associated with endometrial cancer with relative risk of 1.72, a summary relative risk of 1.56 in 9 of the cohort studies, and 1.85 in 13 case control studies [13]. With respect to diabetes-attributed mortality rate among women with endometrial cancer, reported statistics highlighted that the risk of disease-specific mortality was 32% times higher among women with versus without diabetes [10, 15].

The relationship between diabetes and endometrial cancer may be attributed to shared biologic factors such as obesity and age [16, 17]. In this respect, obesity has been identified as a major risk factor for both diseases [16, 17], due to its association with insulin resistance that results in type 2 diabetes [17]. Obesity is also linked to the excess body fat that leads to hormonal imbalances and elevation in the levels of estrogen unopposed by progesterone which can then lead to early stages of endometrial cancer formation [13, 14, 18]. With respect to age, it was established that older age is a risk factor for both diabetes and endometrial cancer [5, 16]. In this regard, recent statistics indicated that the incidence and mortality rates of endometrial cancer increase with age [5], particularly among women aged 55 or more [19], with approximately 80% of new cases and 91.3% of endometrial cancer-specific deaths in the US occurring within this age group [19]. The effect of age on endometrial cancer may be ascribed to the hormone replacement therapy. In specific, estrogen-only replacement therapy [20], that is recommended for women going through the menopausal transition typically occurring between the ages of 45 and 55 [21], was shown to double the risk of endometrial cancer [22].

Behavioral factors, such as cigarette smoking and alcohol consumption, were also identified as factors that may affect endometrial cancer. In this regard, some epidemiological studies have determined an association between smoking and a decreased risk of endometrial cancer [23‐25], possibly due to its anti-estrogenic effect through altering hormone metabolism [23] and consequent weight loss [23, 24]. Emerging evidence has also suggested that there may be a potential relationship between alcohol consumption and endometrial cancer [26‐28]. This relationship may be attributed to the fact that alcohol can raise the levels of estrogen in the blood which then increases the risk of endometrial cancer [27].

In conjunction with the aforementioned biologic and behavioral factors, racial and social disparities in the distribution of endometrial cancer have been marked across the different socioeconomic classes. More specifically, various indices of socioeconomic status, referred to as social determinants of health (SDOH), including marital status, level of education, and healthcare coverage, were shown to have an effect on endometrial cancer [29‐32]. For instance, unavailability of healthcare coverage and lower levels of education were identified as risk factors for endometrial cancer in several studies [29‐31]. Specifically, women with no healthcare coverage were shown to have higher rates of advanced-stage diagnosis of endometrial cancer and unequal access to treatment compared to women with healthcare coverage [31]. Conversely, being married was associated with lower mortality rates compared to being divorced (Hazard Ratio HR 1.19), widowed (HR 1.22), or never married (HR 1.23) [32].

Disparities in endometrial cancer were also highlighted among women of different races whereby White women are more frequently diagnosed with endometrial cancer compared to women of other races [33]. However, Black women tend to have more advanced stages and aggressive tumors of endometrial cancer compared to White women [33, 34]. Similarly, the risk of death attributed to endometrial cancer among Black women was shown to be 2.5 times higher than that of White women, and the disease-specific mortality rate of endometrial cancer among women of Black race was reported as 9.2 per 100,000 compared to 4.6 per 100,000 for women of White race [35]. These statistics reflect a higher prevalence in the incidence of endometrial cancer among White women, but increased mortality rates among Black females. The aforementioned racial disparities in endometrial cancer incidence and disease-specific mortality rates may be credited to embedded differences in the exposure to risk factors, comorbidities, and unequal access to health care providers, diagnosis, and treatment services among the different racial groups [31].

Despite that the effects of all of these social, biologic, and behavioral determinants of health were independently reported on diabetes and cancer, no study has offered a comprehensive understanding of the collective effect of these determinants of health, along with diabetes on the risk of endometrial cancer among women. Therefore, the objective of this study is to address this gap in knowledge by investigating, for the first time, the association between this cluster of predictors and endometrial cancer using a nationally representative sample of American women adopted from the Behavioral Risk Factor Surveillance System (BRFSS). We hypothesize that the occurrence and development of endometrial cancer are increased by diabetes and indices of social determinants of health. The acquirement of this knowledge is important in order to develop new strategies to alleviate the burden of endometrial cancer, by addressing the contribution of biologic, social and racial determinants on the incidence, development and progression of endometrial cancer.

In the present study, we examined the cumulative effect of a cluster of covariates that included diabetes, SDOH, behavioral, and biologic factors on endometrial cancer among women in the US. Our findings indicated that diabetes, biologic factors (age and obesity) and social and racial determinants are associated with the risk of occurrence and development of endometrial cancer.

Our analysis revealed that women with diabetes had almost double the estimated risk of endometrial cancer compared to women without diabetes. This detected association could be attributed to hyperinsulinemia in type 2 diabetes that contributes to elevated estrogen levels and consequently the development of endometrial cancer [5, 9]. Our result is consistent with previous findings which reported an increased risk of endometrial cancer among diabetic women with risk ratios ranging between 1.7 and 2.1 [7‐15].

Our analysis also revealed significant associations between biologic factors (age and obesity), as well as indices of social determinants of health (SDOH), with this type of cancer. However, it did not detect significant associations between behavioral factors (smoking and heavy alcohol consumption) and endometrial cancer. The undetected association is not surprising since the literature had inconsistent findings concerning the effect of smoking and heavy alcohol consumption on the risk of endometrial cancer [23, 26‐28, 40]. Our results indicated that older age and obesity were more prevalent among women with diabetes and were significantly associated with increased odds of endometrial cancer. Among the SDOH, our data indicated that the level of education of attending college or technical school was also associated with increased odds of endometrial cancer and a higher prevalence of this level of education among women with diabetes. This observation can be driven by the fact that people with lower levels of education tend to miss more on preventative checkup visits and available screening facilities [41], which may contribute to under-reporting of cases among individuals with lower levels of education compared to those who are more educated.

Conversely, individuals who rented their homes, had other living arrangements, were divorced, had incomes over $35,000, and were of Black or other race had lower odds of endometrial cancer and were less prevalent among women with diabetes compared to their respective reference categories. However, the result pertinent to Black women need to be carefully interpreted in view of the uterine cancer high mortality rate in this group of women [42]. In particular, Black women were shown to be more prone of diagnosis with more aggressive forms of endometrial cancer compared to other races, resulting in an increased disease-attributed mortality rate among this group of women [42]. This may lead to an under-representation of Black women in the BRFSS study that are missed either due to the severity of their endometrial cancer related illness, or death.

Accordingly, our results highlighted the relationship between diabetes, biologic and social determinants of health, with endometrial cancer, and indicated that the discrepancies in the diagnosis of diabetes and the incidence of endometrial cancer may be related to racial and socioeconomic differences.

One potential explanation for the effect of age and obesity on endometrial cancer could be related to menopause and high levels of body fat. Menopause, a physiological change that typically occurs among women of older age, is known to be associated with the growth of the tissues in the lining of the uterus into its muscular wall, increasing the risk of endometrial cancer [3]. In addition, hormone replacement therapies (unopposed estrogen, combined estrogen and progesterone, or tibolone) used by women having menopausal symptoms [20] were shown to double the risk of endometrial cancer [22]. Fatty tissues that result from obesity promote higher levels of estrogen which contribute to the development of endometrial cancer [5]. In addition, high levels of fat in the abdomen observed during menopausal years, combined with older age, can also increase the risk of endometrial cancer [43].

Furthermore, the observed social and racial disparities in endometrial cancer may be linked to differences in healthcare access, screening facilities, and quality of care [5] which ultimately reflect on the detection and reporting of new cancer cases in the different socioeconomic and racial groups. For instance, compared to White women, Black women were found to have less healthcare coverage and limited access to preventive medical care such as screening and genetic testing [5]. Married women were reported to have earlier diagnoses with better prognoses for this type of cancer and greater compliance with regular medical checkups in general, compared to non-married women [32]. This can due to the mental, social, and financial support provided by their spouses [32, 44‐46]. Awareness of the symptoms and risk factors for this disease, and its early diagnosis, are usually coupled with higher levels of education and financial stability that enable women to lead a healthier lifestyle and have better access to preventive measures and quality of care [30, 47‐49].

Despite the novelty of the data presented in our study, there are some limitations that need to be highlighted. First, given the cross-sectional nature of the BRFSS, all the detected relationships should be interpreted as associations and not as causal-effect inferences. Moreover, the odds ratios reported in this study can be viewed as prevalence odds ratio since endometrial cancer might have occurred before diabetes. However, this was not a major concern in our study since only very few women were diagnosed with endometrial cancer before their diagnosis with diabetes. This issue was addressed in the additional analysis that we performed in which we excluded this group of women, and the results aligned with all the associations that were detected in our original analysis. The study population of the BRFSS may be more representative of women who survived their endometrial cancer, and less reflective of women with more aggressive forms of this disease. The latter group may have been under-represented either because women were too ill to respond or have passed away due to endometrial cancer related mortality. Thus, this surveillance system might have missed on women who were diagnosed further in the past and subsequently died, and may have captured more, women with less severe past diagnosis of endometrial cancer. Missing out on potential severe cases may bias our results and study population towards more survivorship than the overall general population of endometrial cancer.

In response to the survey, 22 states agreed to administer the questions on cancer survivorship, while the remaining states declined to participate in this part of the questionnaire. This for sure imposes some limitation on the generalizability of the results. In addition, our data had only 106 women who had both diabetes and endometrial cancer, which can also pose some limitation on the inferences. However, the fact that our results showed a consistent association between diabetes and endometrial cancer, that continued to prevail despite of the large denominator of 84,118 female participants, comes in support of the presence of a strong relationship between these two comorbid diseases. Lastly, it is important to note that the BRFSS data might be subject to recall and social desirability bias due to the self-reported information. Nonetheless, validity of this data was underscored in the context of studies where self-reporting entailed easy-to-understood questions, secured anonymity, and full absence of reprisal [50]. These settings mirror the context of the data collection process that was undertaken in the BRFSS study.

In conclusion, our study is the first to adopt a comprehensive approach to the assessment of the effect of diabetes, SDOH, biologic and behavioral factors on endometrial cancer using nationally representative data. In this study we provided further evidence that underlie the growing burden of diabetes with increased risk for endometrial cancer progression. Our data also point to the social and racial disparities associated with poor prognosis of women with endometrial cancer. In light of our data and with the increasing incidence rates, endometrial cancer is set to become a significant public health problem [7] which necessitates corrective measures at the level of modifiable risk factors such as the ones we addressed in our study. The observed social disparities in the health outcomes of endometrial cancer can be reconciled by adopting policies and social structures that endorse standardized early detection management programs and preventative strategies covered by the healthcare system and advocate for equitable access to healthcare services. Active implementation of such recommendations is key for addressing social and racial inequities in health and reducing the burden of this cancer among women. Future studies could further explore the mechanisms through which diabetes, and social determinants of health modulate biologic markers on risk to develop endometrial cancer.