Up to now, the studies involving diabetes mellitus and malignancies show controversial results: Many of them have found incidences of malignancies that were comparable or even lower than those in nondiabetic subjects; others conclude that diabetes mellitus is linked to a higher incidence of malignancies and/or a predictor of mortality from cancer. Insulin and its precursors, pro- and pre-proinsulin, have been shown to have some homology to the insulin-like growth factors, but, moreover they have some affinity to bind at receptors of the tumor growth factor and some hybrids too. Hence, an association between diabetes mellitus, insulin, hyperinsulinaemia, and carcinogenesis appears plausible. On the other hand, diabetes mellitus can influence different hormone levels. In some tumor entities, such as prostate carcinoma, this effect can somewhat counterbalance the direct mitogene effect of insulin and its precursors. All in all, as a result of the complexity of these mechanisms and the differences between the tumor entities, the question whether diabetes mellitus is associated with an increased or a reduced risk for the development and in respect of the prognosis of cancer cannot be answered. The only way to give some answer is to focus on specific tumor entities: It seems that diabetes mellitus and/or hyperglycaemia are independent risk factors and/or predictors at least in respect of cancer of the colon, pancreas, female breast, endometrium, and, in men, of the liver and bladder. However, most of these data were assessed in patients with type 2 diabetes mellitus. This makes it highly questionable whether the data can easily be transferred to patients with type 1 diabetes. Moreover, additional potential limitations are that most of the studies do not focus on the treatment modality or the race of participants. In conclusion, up to the present, we have an increased risk for some and a reduced risk for other tumor entities, but still, we cannot give the general answer.
