Diabetes and cancer
The association between carcinoma of the pancreas or endometrium and diabetes has been known for many years; associations between type 2 diabetes and a range of other cancers including breast, colorectum and bladder have come to light more recently. The association between type 1 diabetes and cancer remains more speculative. Type 2 diabetes is associated with obesity, a major risk factor for cancer, and diabetes and obesity predispose to the same range of tumors; hyperglycemia as such does not appear to be a major risk factor. Hyperinsulinemia and/or increased exposure to the insulin-like growth factors may provide the link between diabetes, obesity and cancer. There is controversy as to whether therapies used for diabetes influence the risk of cancer. Cancer mortality is increased in people with diabetes, but recent studies suggest that this excess is due to the diabetes rather than to the neoplasm. Future morbidity and mortality from cancer in the diabetic population will increase in proportion to our success in controlling cardiovascular risk.
An assocation between diabetes and Pancreatic cancer has been recognised for more than a century, and the association with Endometrial cancer dates back to the middle of the century. Elliot Joslin noted that people with diabetes were less likely to die of cancer, mainly because they were more likely to die from vascular disease.
Perhaps for this reason, diabetologists paid little attention to cancer until reports emerged suggesting an association between a range of Glucose-lowering therapies and cancer. Cancer investigators meanwhile had long been investigating the relation between obesity, diabetes, the metabolic syndrome and cancer, and had gone a long way towards establishing hyperinsulinemia and/or insulin resistance as a major factor or factors in its pathogenesis. The resultant meeting of investigators from different background proved very fruitful, although methodological problems initially caused mistakes to be made and false trails to be pursued.
Many epidemiological studies have shown an association between type 2 diabetes and a range of cancer types. The strongest associations have been seen with primary liver cancer and carcinoma of the pancreas, although there may be an element of reverse causality (i.e. the cancer promotes or aggravates diabetes). The risk of endometrial cancer is doubled in women, and risks of cancer of the breast, colorectum and bladder, together with non-Hodgkin’s lymphoma, are increased by 20-40%. In contrast, the risk of prostate cancer is reduced by around 15%. There is no consistent association with cancer of the lung and ovary, and no strong association with other types of cancer has emerged.
Cancer is most likely to be diagnosed in the years following diagnosis of diabetes or conversion to insulin, possibly due to detection bias (i.e. a cancer is more likely to be detected in those attending doctors for other reasons). Alternatively, this might in some cases be due to reverse causality (i.e. tumor-induced worsening of diabetes). People with diabetes have a higher mortality for some types of cancer, probably due to higher mortality from diabetes-related causes rather than neoplasia itself.
Obesity is a major acquired risk factor for cancer, is strongly associated with type 2 diabetes, and confers similar risks for a similar range of tumors. In contrast, hyperglycemia, the hallmark of diabetes, does not appear to be a major risk factor. Thus, cancer risk is not associated with the duration or intensity of hyperglycemic exposure, and glucose-lowering therapy does not appear to diminish the risk of cancer in diabetes (see Hyperglycemia and cancer).
Insulin has long been known to be mitogenic (i.e. it stimulates cell growth in vitro), but is not mutagenic (i.e. does not transform healthy cells into cancer cells). Raised levels of insulin and the insulin-like growth factors (IGFs) are strongly associated with increased cancer risk, and might explain the increased risk observed in obesity and type 2 diabetes (see Hyperinsulinemia and cancer)
Diabetes Therapy and Cancer
Metformin and cancer risk. Animal studies and theoretical considerations initially suggested that metformin might be protective against certain forms of cancer, and remarkably positive findings emerged from a series of pharmacoepidemiological studies. Unfortunately, methodological errors including immortal time bias may have invalidated many of these investigations, and appropriately performed observational studies and pooled analysis of randomized controlled trial data suggest that metformin does not influence cancer risk at therapeutic levels for diabetes.
Insulin analogs and cancer. One of the early insulin analogs, AspB10 insulin, was withdrawn from development because of the development of mammary tumors in rats. Insulin glargine, a widely used-long acting analog, is considerably more mitogenic in vitro than human insulin, although it rapidly dissociates into non-mitogenic metabolites following injection. Although it was rapidly established that overall cancer risk is not increased with this insulin, several studies suggested an association with post-menopausal breast cancer. Although this association has not been entirely excluded, current opinion suggests that any increase in the risk of breast cancer is either non-existent or very small.
TZDs and cancer. Several lines of evidence suggest that thiazolidinediones may modulate the risk of cancer either positively or negatively, but the observation that has attracted most interest is that pioglitazone has been associated with a moderate increase in the risk of bladder cancer, and has been withdrawn from the market in France and Germany for this reason.
GLP-1 based therapies and cancer. Both classes of incretin therapy (GLP-1 agonists and DPP4 inhibitors) have been controversially associated with a two-fold risk of acute pancreatitis. GLP-1 receptors are present on exocrine pancreatic duct cells and sustained exposure to such therapies has been reported to produce duct hypertrophy and pancreatic enlargement in species including humans. The concern has been raised that subclinical inflammation associated with these changes might predispose to pancreatic cancer, for which a signal exists in two regulatory databases. Alpha cell proliferation and neuroendocrine pancreatic tumor formation is a further concern.
Thyroid C-cell tumors are reported in rats, and GLP-1 receptors are also expressed in human thyroid and 20% of medullary carcinomas of the thyroid, and this is also under review.
^ Giovannucci E et al. Diabetes and cancer. A consensus report. Diabetes Care 2010;33:1674-85
^ Johnson JA et al. Diabetes and cancer (1). Evaluating the temporal relationship between type 2 diabetes and cancer. Diabetologia 2012;55:1607-18
^ Renehan AG et al. Diabetes and cancer. (2)Evaluating the impact of diabetes on mortality in cancer. Diabetologia 2012;55:1619-32
^ Renehan AG et al. Body mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 2008;371:569-78
^ Suissa S, Azoulay L. Metformin and the risk of cancer: time-related biases in observational studies. Diabetes Care 2012;35:2665-73
^ Butler AE et al. Marked expansion of exocrine and endocrine pancreas with incretin therapy in humans with increased pancreatic dysplasia and the potetial for glucagon-producing neuroendocrine tumors. Diabetes 22 Mar [Epub ahead of print]