This article is more than 5 years old. Information may no longer be current.
Cancer incidence in type 1 diabetes similar to general population
Site-specific cancer patterns in type 1 diabetes largely follow incidence in patients with type 2 diabetes, but more minimal, and do not drastically differ from general populations, according to a presentation at the 50th European Association for the Study of Diabetes Annual Meeting.
A multicenter, multinational study showed higher rates of stomach, colorectal, liver, pancreas, kidney, thyroid and endometrium cancers and lower rates of prostate and testis cancers, as well as breast and melanoma cancers but only in the short term, according to the presentation.
“All of the previous studies have been based on all diabetes patients, so that is mainly results for type 2 diabetes,” said Bendix Carstensen, MSc, of Steno Diabetes Center in Gentofte, Denmark. “There is no overall excess cancer among type 1 diabetes patients. We can exclude a major carcinogenic effect of exogenous insulin in type 1 diabetes patients.”
The main reason such studies have not been done to date is because type 1 diabetes status was not well defined in older databases since the definition is relatively recent, Carstensen said; therefore, follow-up is inherently in younger ages with less cancer incidence.
Carstensen and colleagues from other institutions investigated nationwide diabetes registers from Australia (1997-2008), Denmark (1995-2012), Finland (1972-2010), Scotland (1995-2011) and Sweden (1987-2011) to determine cancer occurrence in the population with type 1 diabetes.
The disease was defined in three categories — as diagnosis before age 30 years, age 35 years or age 40 years — for sensitivity and showed “no difference whatsoever in the results,” Carstensen said; the data uses age 40 years as a cut point.
The researchers compared diabetes cancer incidence rates with population cancer incidence rates from nationwide cancer registries. Poisson models were used, with adjustments for age, date of follow-up and date of birth.
The investigators estimated the overall incidence risk ratio for all patients with type 1 diabetes, as well as the effect of time since diabetes diagnosis.
During 4.6 million person-years of follow-up, 9,400 cancer cases were identified among patients with type 1 diabetes — 1,337 cases in Australia; 1,042 in Denmark; 2,408 in Finland; 533 in Scotland; and 4,049 in Sweden. The median age at cancer diagnosis was 51.1 years (interquartile range, 43.5-59.5). Among men, risk ratio for any type of cancer was 1.00 (95% CI, 0.97-1.03) compared with 1.05 (95% CI, 1.02-1.08) among women.
“Overall, there’s no excess risk among men and small excess risk among women,” Carstensen said. “There is about 20% increase for stomach cancer for men and 70% for women.”
Men and women with type 1 diabetes were at higher risk, respectively, for stomach cancer (RR=1.19 and 1.75), colorectal cancer (RR=1.13 and 1.09), liver cancer (RR=2.18 and 1.63), pancreas cancer (RR=1.70 and 1.36), endometrial cancer (RR=1.53), kidney cancer (RR=1.29 and 1.42) and thyroid cancer (RR=1.27 and 1.48); all risks were significant. Patients demonstrated decreased risk for cancer of the prostate, testis, breast and melanoma.
Diabetes duration demonstrated an effect, with risks greatest during the first year (RR=2.5), then decreasing between 2 and 5 years for both men (RR=1.2) and women (RR=1.1).
“There is a substantially higher rate ratio in the very first period after diagnosis, which can be ascribed to ascertainment bias and increased medical surveillance,” Carstensen said.
For more information:
Carstensen B. Abstract #245. Presented at: 50th EASD Annual Meeting; Sept. 16-19, 2014; Vienna.
Disclosure: The research was supported by the European Foundation for the Study of Diabetes.
Perspective
Back to Top
A possible increase in cancer risk among patients with diabetes has long been an area of concern and study. Insulin is a potent growth factor, making hyperinsulinemia a potential cause of abnormal cell growth. For patients with type 1 diabetes, there have also been retrospective studies linking insulin analogs to increased cancer risk. Fortunately, additional studies have demonstrated the overall safety of insulin analogs. With this background in mind, the researchers looked at large clinical databases and found that, overall, there was not increase in cancer risk among people with Type 1 diabetes when compared with a population without diabetes. This finding is not surprising, but still reassuring. The use of multi-year registries from five countries adds to the validity of the study. There did appear to be some malignancies that were more (or less) frequent in patients with Type 1 diabetes, and it is difficult to know how to interpret those findings. Nevertheless, overall risk was not increased, and to be able to answer “no” answer to the question “Am I more likely to get cancer because I have type 1 diabetes?” is a welcome finding.
Perspective
Back to Top
This study is very interesting because we have long been aware that there appears to be a link between type 2 diabetes and cancer risk. Whether the cancer risk is really related to diabetes remains a bit uncertain because, as the authors point out, people with newly diagnosed diabetes may be accessing medical care for the first time or more frequently than before their diagnosis. As a result, pre-existing cancers may come to the forefront, in a way they wouldn’t have if they hadn’t had the diabetes diagnosed.
The use of exogenous insulin is one potential mechanism though which diabetes could increase cancer risk. In this study, the authors looked at the incidence of cancer in patients with type 1 diabetes, all of whom would be on exogenous insulin, and wondered whether they would have the same cancer profile as patients with type 2 diabetes.
This has really never been studied before, so this is one of the strengths of the report. The authors also looked at registries from five different countries around the world. They found essentially the same pattern of cancer as seen in type 2: an increased risk for pancreatic and liver cancer, and a decreased risk for prostate cancer.
This tells us that diabetes itself — whether it be type 1 or type 2 — rather than exogenous insulin that seems to increase cancer risk.
The limitation, which the authors clearly admit, is their diagnosis of type 1 is based on age; just because a patient is younger at diagnosis does not mean he or she has type 1 diabetes. However, the study still provides some evidence there’s a risk of cancer related to diabetes that we don’t completely understand yet and more work needs to be done in that area.