Diabetes mellitus (DM), a possible adverse effect of childhood cancer treatment, is strongly associated with cardiovascular disease and early mortality in adult childhood cancer survivors (CCS). Here, we assess the prevalence and determinants of DM in our nationwide CCS cohort. In this cross-sectional study, the prevalence of DM was assessed in 2,338 CCS, using the Lifelines cohort (n=132,226 adults with no history of cancer) as a reference. DM was defined through serum glucose measurement (fasting glucose ≥7.0mmol/L or non-fasting ≥11.1mmol/L) and/or self-report (previous diagnosis and/or medication use). Multivariable logistic regression models, adjusted for age, sex, and BMI, were used to assess the cohort effect on the presence of DM. Multivariable logistic regression analysis was used to identify determinants of DM in CCS. Survivors (median age 34.7 years, median follow-up time 27.1 years) showed increased odds for hyperglycemia (aOR=2.72, 95%CI=2.06-3.59), previous DM diagnosis (aOR=3.03, 95%CI=2.33-3.95), and anti-diabetic medication use (aOR=2.94, 95%CI=2.17-3.99), compared to the reference cohort. Age (OR=4.32, 95%CI=1.84-10.15, >35 versus 18-35 years), BMI (OR=1.12, 95%CI=1.08-1.16, per point), family history of DM (OR=2.38, 95%CI=1.51-3.76), prior abdominal/pelvic radiotherapy (OR=4.19, 95%CI=2.32-7.55), total body irradiation (OR=14.31, 95%CI=6.98-29.34), hypogonadism (OR=2.40, 95%CI=1.15-4.99), hypertension (OR=1.71, 95%CI=1.06-2.76), and dyslipidemia (OR=3.81, 95%CI=2.15-6.75) were associated with DM in CCS. A significant interaction between age and sex on the development of DM in survivors was identified. The identified three-fold increased risk of DM in CCS, along with the clinically relevant -and some modifiable- determinants, underscores the importance of early risk-based screening and the exploration of lifestyle interventions in this population.
Read full abstract