Dissimilar Associations of Same Metabolic Parameters with Main Chronic Noncommunicable Diseases (Cancer vs Some Other NCDs)

Summary

Hormone-dependent tissues’ cancers (mainly breast and endometrial and several others) are among the most frequent malignancies in adults and are often discussed in context of their correlation with other chronic noncommunicable diseases (NCDs), for example, cardiovascular and cerebrovascular conditions, and their risk factors, which may also be hormone metabolic. An idea that is often expressed delineates common factors leading to NCDs of malignant and nonmalignant nature. However, this idea is not always confirmed by study results. The reasons for this discrepancy are not clear and require further analysis. This editorial tries to show the importance of this problem with a few examples (in particular, by attracting information on the role of birthweight, adult height and family history of diabetes) which may help us understand some mechanisms behind interconnections of major NCDs, including cancer.

Keywords::

• adult height
• birthweight
• body fat/obesity
• cancer
• cardiovascular diseases
• cerebrovascular diseases
• family history of diabetes Type 2
• stroke

According to WHO, the four main types of chronic noncommunicable diseases (NCDs) are cancer, cardiovascular and cerebrovascular conditions (e.g., heart attacks and stroke), chronic respiratory diseases (such as chronic obstructive pulmonary disease) and diabetes [1]. NCDs are by far the leading cause of death in the world, representing more than 60% of all annual deaths; in 2013, NCDs caused the deaths of almost 37 million people per year [1].

In this paper, we aim to confine ourselves to the comparison of cancers in hormone-dependent tissues with cardiovascular and cerebrovascular conditions, trying to concentrate on their relationship with metabolic risk factors, including diabetes and some others.

Despite being a historically well-discussed topic, this issue continues to be a point of interest. There are some recent publications in major journals (e.g., DJ Hunter and KS Reddy [2]), along with a well-known joint agenda of the three societies (the American Cancer Society, the American Diabetes Association and the American Heart Association), where the accent was made on common risk factors for cancer, cardiovascular disease and diabetes and on the use of rather similar approaches to their prevention [3].

In this editorial, we are going to discuss several examples which show that, in spite of being a milestone in research, the concept of common hormone-metabolic risk factors is not universally true. We will also try to find out if there are any ‘dissimilar associations’ between these factors and NCDs and what kind of associations these may be.

Example 1: birthweight

A newborn’s birthweight, mentioned in the context of NCD, is most invariably associated with the concept of fetal or prenatal programming. It implies the crucial role of the fetal environment and of prenatal exposures in forming postnatal predispositions to different conditions, which become drastically important in the second half of a patient’s life, when the incidence of NCD peaks [4,5]. In this case, however, we would also like to mention another observation according to which the child’s birthweight may predict not only the child’s own predispositions (which is rather well known) but their mother’s health-related conditions too [6,7].

The data suggest a certain ‘dichotomy’ in newborn birthweight whereby the mother and child have different propensities for developing NCDs. Thus, mothers giving birth to smaller children (<2500 g) have a higher risk of cardiovascular diseases [7,8]. Contrariwise, the maternal cancer risk (mostly hormone-dependent malignancies like breast and endometrial cancer) is higher in women giving birth to larger babies weighting ≥4000 g [6,7,9]. These ‘nonstandard’ facts are likely to be particularly important if we consider that, according to some data, cardiovascular diseases (mostly ischemic heart disease) and hormone-dependent tumors tend to affect women of the same age group (mostly ≥50 years old), who are also more likely to be exposed to the same risk factors (e.g., high saturated fat consumption) and are inclined to have similar hormone-metabolic disorders such as carbohydrate metabolism disturbances and obesity [3].

However, an exact reason for these differences in both cancer and noncancer maternal morbidity, in relation to birthweight of their children, is still an open issue. First of all, not all the cases studied provide an expected hormone-metabolic profile or the associated structural changes. Thus, signs of atherosclerosis, which play a role in the development of cardiovascular and cerebrovascular conditions, do not always accompany hormone-dependent tissues’ cancer development. There may be even an inverse correlation between the incidence of these conditions as was demonstrated in a historic cohort 20 years ago [10] and then again more recently [11].

Moreover, women who are affected several decades after becoming a mother by cancer, ischemic heart disease or diabetes are different from those affected by stroke not only in the child’s birthweight history but also in the child’s weight gain dynamics in early childhood [12,13]. Furthermore, other parameters (besides fat mass) can influence body weight in both children and adults, such as muscle mass and height [14]. The latter parameter, in the context of cancer and noncancer morbidity in adult life, is discussed in more detail below.

Besides, there are data which allow us to suppose the existence of a certain degree of competition (or trade-off) between different NCDs based on the concept of multiple death causes [15].

This mostly concerns the possibility of the competing risks associated with cardiovascular conditions and hormone-dependent (breast, endometrium, thyroid gland, prostate, etc.) or hormone-independent tissues’ cancer. Competing risk is a common idea in mothers who give birth to particularly large or small children, although there are still many other discrepancies to consider. For instance, the presence of accompanying conditions, such as Type 2 diabetes (T2D) [16], which may be either sporadic or associated with family trait, that will be discussed further. The individual characteristics of endocrine status dynamics (in particular, hormone and growth factor levels) during onto- and oncogenesis in different patients may also be an important, although not so universal, reason for these aforementioned discrepancies. In addition, some ethnic (race-related) and exogenic factors (like smoking, etc.) also play a significant role in both cancer- and noncancer-associated mortality of females having children, since many of these factors are connected with NCD structure in later life or with NCDs’ more or less aggressive clinical course [9]. With regard to cancer, the influence of several factors that have been mentioned may be explained by the specific mechanisms of hormonal carcinogenesis [17,18] as well as by certain characteristics of the diseased individuals.

Example 2: adult height

Although obesity and fat mass, as well as insulin resistance and metabolic syndrome, are most often mentioned as possible causative factors for some malignant tumors and several other NCDs, another characteristic, namely, a person’s height in adult life, can be considered a causative factor too. Although the anthropometric parameters have long been studied as potential risk factors for the development of NCDs, only recently did a large-cohort study of more than a million people (both male and female) underline the positive correlation between adult height and cancer-related death (most studied localizations) and negative correlation between height and death from ischemic heart disease or cerebrovascular incidents. The association between adult height variations and diverse effects on several major adult-onset diseases was one of the study’s main conclusions [19].

There are now several publications that help us understand the role of height values, such as that issued by German and American researchers headed by N Stefan and M Schulze [14]. Scrutinizing the problem from entirely different points of view, these authors state that, although considering BMI and waist circumference as being surrogate markers of excessive visceral fat content are often viewed as a ‘signal’ trait associated with higher risk of main NCDs, much less attention is devoted to such a parameter as adult height. Meanwhile, according to accumulated data, the average height is increasing worldwide under the process of acceleration as well as a result of changes in external factors, particularly dietary habits. For instance, an increased consumption of dairy products can not only influence a child’s growth, but may also be a cause for earlier menarche, which is a risk factor in breast cancer [20].

Furthermore, the final height of an individual depends significantly on the effect of IGF-1 and IGF-1’s interaction with its binding proteins. IGF-1 activity influences cellular proliferation processes involved in leg bone elongation during growth periods [14,21,22], which also distinguishes height from other NCD-associated characteristics [21,22] since the high IGF-1 levels promote cellular hyperplasia in so-called ‘target tissues’. Due to this, increased proliferation is often recognized as a local marker of predisposition to cancer and its risk factor. On the other hand, tall people less frequently show signs of insulin resistance, with one study showing an inverse correlation between leg length value and T2D incidence [23].

Notably, in nonobese people with incident (newly identified) diabetes, the diabetes-associated mortality rate is actually higher compared with the matched obese cohort. This is notable since noncardiovascular mortality (including cancer) of nonobese people with newly diagnosed diabetes was 2.25-times higher than in an analogous group of obese individuals, while cardiovascular mortality was simultaneously only 1.47-times higher when compared with people with excessive body weight [23]. We should also note that, according to data published by Furer et al. [24], short stature in men at young adulthood is associated with a higher incidence of newly diagnosed diabetes. Thus, at the age of 30 years, each 1-cm decrement in adult height was associated with a 2.5% increase in diabetes adjusted risk (HR: 1.025; 95% CI: 1.01–1.04; p = 0.001). This trend is further augmented by the presence of a family history of diabetes [24], which is also an important issue in the context of NCDs.

Example 3: family history of diabetes

There is no real doubt that diabetes needs to be considered as a socially significant pathology, especially in light of accumulating data on the increase in its prevalence reaching an epidemic scale [25]. From an oncologist’s point of view, T2D is the most important as it accounts for around 90% of all diabetes cases. The most commonly discussed T2D risk factors are excessive body weight, lack of exercise, diet, age and a previous family history of diabetes. Although the hereditary nature of diabetes has been studied for many years, the precise effect of anthropometric parameters, lifestyle and, most importantly, genetic factors, in transferring the condition to subsequent generations is still considered to be relatively minor [26]. Therefore, in this discussion, we are going to omit a clearly inherited form of diabetes, maturity onset diabetes of the young. Maturity onset diabetes of the young, which is rare (<1% of diabetes mellitus cases) and mostly manifests in young people aged less than 25–30 years, is characterized by distinct genetic mutations and may develop in individuals without excessive weight [27].

A significant history of diabetes within a family results not only in a higher T2D risk but also increases the rate of prediabetes, gestational diabetes and obesity. However, while a family history of diabetes increases T2D incidence and consequently should have done the same for cancer morbidity, there are some observations stating quite the opposite; namely, there is a tendency for fewer cancer cases to be registered in a cohort of diabetics with a family history of diabetes [28–30]. This may be explained by the effect of certain antidiabetes medications prescribed to T2D patients (in particular, metformin) [29] as well as by the affiliation of the neoplasms to one or another of their subvariants. Thus, according to our data, when the comparison of patients with different molecular-biological types of endometrial cancer was made, the lowest frequency of familial diabetes was discovered in the group with POLE mutations (8.3%) and the highest (17.9%) in patients without the characteristic molecular profile of the tumor tissue (or so called WCMP type of endometrial cancer) [31]. On the other hand, although the researchers failed to establish any connection between family history of diabetes and the incidence of cerebrovascular pathology (e.g., stroke) [32], it was noted that the families affected by diabetes are at higher risk of cardiovascular diseases [33].

All these facts are simply further justification for the idea of a ‘dichotomy’ in the occurrence of common NCDs (of malignant as well as nonmalignant type) since, in real life, they may respond rather differently to the same hormone-metabolic risk factors. This phenomenon certainly needs further analysis, with the aim to find the approaches which allow for optimal action in such unusual (or nonconventional) ‘arrangement’ of different NCDs.

Conclusion

We should stress that, after infection lost its prevalence in modern disease mortality and NCDs started to take over, the issue of a possible predisposing to NCDs factors (and their similarity/dissimilarity) is becoming more and more important.

Besides what has already been discussed of the influence of hormone-metabolic signals, which can affect the rate of cancer and other NCDs differently and manifest in such characteristics as high or short adult stature [14,22], family history of diabetes [28,30] and baby’s low or high birthweight (important not only for the baby but also for the mother [7,8]), there are two additional points to be mentioned.

One of these points may be summarized by the use of the term ‘comorbidity’, under which one can understand additional conditions found in cancer patients that are not only compromising the anticancer therapy effectiveness but could also influence so-called noncancer-specific mortality [34]. In relation to this, we also often see that, although there is a stable increase in NCDs’ incidence with age, this increase is probably not always ‘infinite’ or parallel if evaluated individually (in the same people). Some examples were presented above.

The second point returns us to publications on familial diabetes incidence in diabetics with or without concurrent hormone-dependent tissue cancers [35]. The authors of this paper formed a preliminary Hypothesis of Metabolic Protective Adaptation, according to which, during the shift from the elder/‘trained’ generation of diabetics (parents) to the younger generation of their relatives, which have reached an adult age, some of the latter individuals may acquire characteristics allowing to ameliorate diabetes-associated genotoxic stress. These features are mostly based on cellular and endocrine mechanisms, among others, and may lower the risk of cancer development [35] that, collectively, could be an element for further study in the context of the issues described by this article.

Financial&competing interests disclosure

This work was partly supported by Russian Foundation for Basic Research (grant 18-015-00026) and received by LM Berstein. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

Additional information

Funding

This work was partly supported by Russian Foundation for Basic Research (grant 18-015-00026) and received by LM Berstein. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.