Researchers from Steno Diabetes Center in Denmark have performed the largest registry linkage study to date focused on diabetes, cancer and insulin use. The study followed the total Danish population, currently 5.5 million people, during 13 years assessing the relationship between duration of insulin use and the occurrence of cancer.
“Contrary to what was previously thought, the risk of cancer does not increase with longer use of insulin. We can now take away that worry.” says Daniel Witte, manager of the epidemiological research group at Steno Diabetes Center.
The new findings were presented today in Stockholm at the annual European diabetes conference EASD. The study confirms the well established association between diabetes and cancer. This increased incidence was especially evident in the first years after diagnosis but decreased over time. The increased risk of cancer was most pronounced in people using insulin and the study showed that for these patients the risk was highest at the start of insulin treatment and decreased over time.
From a patient perspective, the excess risk is best quantified by comparing the cancer risk during a period of 10 years in individuals of the same age. As an example, 20.9% of non-diabetic Danish men aged 65 can expect to develop cancer in the next 10 years, compared to 22.3% of men with diabetes without insulin and 23.7% of men using insulin. For women aged 65 the expected rate of cancer is 15.4% during 10 years in the general population, 16.1% for women with diabetes without insulin and 19.5% for women using insulin.
The fact that the increased cancer risk decreases the longer the patient has diabetes strongly indicates that diabetes does not cause cancer in itself. Furthermore, the marked drop in risk with long-term insulin use indicates that other factors than insulin play an important role. Patients with type 2 diabetes who require insulin are more likely to be obese, exercise too little and have an unhealthy diet, which are shared risk factors for the need for insulin and for cancer.
Registry linkage studies cannot separate the effect of shared risk factors and the biological effect of insulin. Therefore, the study cannot prove or disprove a causal link between insulin and cancer.
Source: Steno Diabetes Center