Risk factors for cancer development in type 2 diabetes: A retrospective case-control study

Abstract

BackgroundThe risk of several types of cancer is increased in type 2 diabetes mellitus. The earliest possible diagnosis of cancer – difficult within regular outpatient diabetes care - is of utmost importance for patients’ survival. The aim of this multicenter, retrospective (years 1998–2015), case-control study was to identify risk factors associated with malignancy in subjects with diabetes treated in a typical outpatient setting.MethodsIn the databases of 3 diabetic and 1 primary care clinics 203 patients (115 women) with type 2 diabetes mellitus who developed malignancy while treated for diabetes were identified. The control group consisted of 203 strictly age- and gender matched subjects with type 2 diabetes without cancer. Factors associated with diabetes: disease duration, antidiabetic medications use and metabolic control of diabetes were analyzed. Also other variables: BMI (body mass index), smoking habits, place of residence and comorbidities were included into analysis.ResultsThe most prevalent malignancies in men and women together were breast cancer (20.7 %) and colorectal cancer (16.3 %). HbA1c (hemoglobin A1c) level ≥8.5 %, obesity and insulin treatment in dose-dependent and time-varying manner demonstrated significant association with increased risk of malignancy, while metformin use was associated with a lower risk of cancer. Diabetes duration, comorbidities, smoking habits, place of residence and aspirin use did not show significant association with risk of malignancy.ConclusionsIn the outpatient setting the obese patients with poorly controlled insulin treated type 2 diabetes mellitus should be rigorously assessed towards malignancies, particularly breast cancer in women and colorectal cancer in men.

Antidiabetic medications use and cancer risk in type 2 diabetes

“…Although insulin use, especially in monotherapy seem to be harmful in terms of cancer risk, addition of metformin abolishes its negative effect on malignancy, which was observed both in our study, as well as in several others [7,9,27]. It should also be noted that in our study patients treated with insulin-based regimens had longer diabetes duration, worse metabolic control and were older compared to patients on other therapeutic models, which could significantly influence cancer risk.…”

“…In the study by Li et al the cancer risk was gradually increasing with the diabetes duration [24]. Also in our recently published study the risk of malignancy tended to increase with the duration of diabetes [7]. The impact of poor metabolic control on elevated cancer risk was documented in the meta-analysis by de Beer and Liebenberg [28], and was also observed in our previous study [7].…”

“…Also in our recently published study the risk of malignancy tended to increase with the duration of diabetes [7]. The impact of poor metabolic control on elevated cancer risk was documented in the meta-analysis by de Beer and Liebenberg [28], and was also observed in our previous study [7]. Epidemiological and clinical data indicate also higher cancer prevalence among elderly compared to younger patients [29,30].…”

“…Obviously, a cohort study would be more appropriate for the analyses that we performed. However, we decided to perform this subanalysis of our data to more precisely evaluate associations between anti-diabetic medications use and cancer development found in our main study [7]. Another important limitation of this study is relatively small number of patients in treatment subgroups, which influenced statistical power of our findings.…”

“…The main objective of our retrospective, casecontrol study was to analyze risk factors associated with cancer development among type 2 diabetic patients, which was published elsewhere [7]. The aim of this sub-analysis was to further evaluate the risk of cancer development associated with the treatment models based on the most frequently used antidiabetic agents: metformin, sulfonylurea (SU) derivatives and insulin in a well-defined, longitudinally observed diabetic population in a real life outpatient setting.…”