Home / Resources / Clinical Gems / International Textbook of Diabetes Mellitus, 4th Ed., Excerpt #108: Type 2 Diabetes and Cancer Part 1

International Textbook of Diabetes Mellitus, 4th Ed., Excerpt #108: Type 2 Diabetes and Cancer Part 1

Jan 16, 2018


An association between hyperglycemia, diabetes, and cancer has been recognized for many years. Epidemiologists first noted the association between diabetes and cancer in the early part of the twentieth century, while the association between hyperglycemia and cancer was reported in 1885. At that time, in Europe and North America life expectancy was improving, rates of over-nutrition and under-exercise were increasing, there was a rise in the percentage of people that were overweight, and the incidence of diabetes began to climb [1]. The epidemiologists and statisticians of the time examined the association between diabetes and cancer in different populations to determine if there was truly a relationship between these conditions, or if the association was purely related to increased longevity. Their studies found that although both conditions increased with age, diabetes appeared to increase the risk of cancer, independent of age. Meanwhile, physicians had observed differences in glucose homeostasis in individuals with cancer, and researchers such as Otto Warburg were studying the distinct metabolic properties of cancer cells, namely the generation of energy by the fermentation of glucose [2]. Since these early discoveries, a wealth of research has been conducted in the field of diabetes and cancer. Today, in the setting of the global diabetes and obesity epidemics, understanding the epidemiologic links between diabetes and cancer and determining the mechanisms through which these conditions are linked has become a priority for research, in order to prevent and treat cancers that are more likely to occur in those with diabetes.


Epidemiologic studies continued through the twentieth century, with reports of a link between diabetes and the development of pancreatic, endometrial, breast cancer, and hematologic malignancies. However, until the 1990s much of the epidemiologic findings were inconsistent due to many confounding risk factors that were not accounted for in studies, such as obesity, smoking, and hormonal therapy as well as differences in study design. From the 1990s to present, large prospective cohort studies have been conducted, such as the Cancer Prevention Study II (CPS II), the Physicians’ Health Study and the Women’s Health Initiative (WHI) in the USA, and the European Prospective Study into Cancer. In these studies individuals were recruited, data including the presence or absence of diabetes, body mass index (BMI), smoking status, were collected along with serum samples. Participants were then followed for a number of years to determine whether they developed cancer or died from cancer.These studies reported an increased incidence of cancer in individuals with diabetes, and an increased mortality in those with diabetes who develop cancer.

Although many of these studies did not specifically record diabetes type, in the US of adults with diabetes, 90–95% have type 2 diabetes (T2DM). Therefore, at least 90% of the populations in these studies would have T2DM. In contrast to T2DM, the studies into the risk associated with type 1 diabetes (T1DM) and cancer have been largely inconsistent. One major potential reason for variable results is that the majority of studies use either insulin therapy or age cut-offs to define T1DM. These age cut-offs vary widely from under 18 years of age to under 40 years of age; therefore, individuals with type 2 diabetes are very likely to be included in some of these studies. In addition, these studies are a mixture of cohort and case-control studies, which may also contribute to the variable results. With these limitations in mind, individual studies have reported an increased risk of pancreatic, liver, mouth and pharyngeal, stomach, skin and ovarian cancer, as well as leukemia in those with T1DM. The types of cancers that are associated with T1DM are not consistently reported between the studies, and some studies report no association between diabetes and cancer [3].

While the overall risk of cancer is greater in those with T2DM, several studies and meta-analyses have examined which cancers are specifically affected by diabetes. While the earliest epidemiologic studies demonstrated a significantly increased risk of pancreatic cancer in those with diabetes, it was later noted that there was a close temporal relationship between the onset of diabetes and the diagnosis of pancreatic cancer; in fact the development of hyperglycemia and diabetes may be a manifestation of undiagnosed pancreatic cancer, this effect is known as “reverse-causation.” More recent studies have excluded individuals who develop cancers within 5 years of the start of the study in an attempt to avoid this effect and these studies have demonstrated an increase in the risk of pancreatic among other cancers in those with T2DM[4]. The results of the meta-analyses examining specific cancer sites and T2DM are summarized in Table 21.1. Individuals with diabetes have approximately twice the risk of hepatocellular, endometrial, and pancreatic cancer than those without diabetes [5–7]. The risk of biliary tract and renal cancer appear to be increased by 40% and bladder cancer by approximately 20% in those with diabetes [8–10], while the risk of breast cancer is increased by 20% in women and colorectal cancer is increased by 30% in both sexes [11,12]. The risk of esophageal cancer may be increased in men with diabetes, while the risk of certain brain cancers and gastric cancer may be increased in women with diabetes [13–15]. The most notable exception to the general increased risk of cancer is with prostate cancer; epidemiologic studies and meta-analyses have reported a decreased risk of developing prostate cancer in men with diabetes [16,17]. However, in some studies for prostate cancer, similar to those in breast and colorectal cancer, diabetes has been associated with a greater mortality risk, a greater risk of treatment failure, and higher recurrence rate. The majority of studies have been conducted in the US and Europe; however, the results are not necessarily applicable to other populations (Asian, African, and Hispanic) where the prevalence of diabetes, obesity, and different types of cancer varies (e.g. hepatocellular and esophageal cancer is more common in Asia, and other factors may outweigh any increased risk from diabetes). Studies are now being conducted in different racial and ethnic groups in different countries to determine whether the  risk associated with diabetes is applicable worldwide. As set out in Table 21.1, meta-analyses have shown that individuals from Asia with a history of diabetes have no increased risk of esophageal cancer (in contrast to the European population), while similar to other groups they do have an increased risk of hepatocellular cancer, and in contrast to other ethnic groups may have an increased risk of prostate cancer [7,14,17]. Although there is substantial evidence supporting the link between diabetes and the more common epithelial cancers, there is less information available on the less common cancers [4]. These cancers are more difficult to study in large epidemiologic studies, due to their lower incidence in the general population and therefore longer follow-up and larger studies are needed to determine whether diabetes increases the risk of some of the less common cancers. In the CPS II study of 1,053,831 individuals after 26 years of follow-up, increases in mortality from less common cancers such as male breast cancer (fourfold increased risk) and oral and pharyngeal cancers (40% increased risk) were observed in those with diabetes. It should be noted that many studies report an increased risk of mortality in patients with cancer who have diabetes, but any individual with diabetes has almost doubled the age-adjusted mortality of someone without diabetes.Therefore, whether diabetes increases cancer-specific mortality because of more aggressive cancer growth and spread, or whether the increased mortality is merely a reflection of the overall increased mortality seen with diabetes remains unclear [4].

Click here to view all Chapter 21 references.