Abstract
BACKGROUND: The association of pre-diabetes and type 2 diabetes (T2D) with incident lung cancer is uncertain, and the incident risk across the glycemic spectrum is unclear. We aimed to explore the associations of glycosylated hemoglobin (HbA1c), pre-diabetes, and T2D with incident lung cancer in a large prospective cohort.
METHODS: Leveraging a total of 210,779 cancer-free adults recruited in the UK Biobank between 2006 and 2010. We performed multivariable Cox proportional hazards models and restricted cubic spline methods to estimate hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the associations of HbA1c, pre-diabetes, and T2D with incident lung cancer.
RESULTS: During a median follow-up of 11.06 years, 1738 incident lung cancer cases were ascertained. The incidence of lung cancer was 20% higher among people with diabetes (HR: 1.20, 95% CI: 1.02 to 1.42) and 38% higher among people with pre-diabetes (HR: 1.38, 95% CI: 1.15 to 1.65). After dividing people with diabetes by whether taking antidiabetic medications, the incidence was 28% higher among people with diabetes without medications (HR: 1.28, 95% CI: 1.02 to 1.61) and 15% higher among people with diabetes with medications (HR: 1.15, 95% CI: 0.93 to 1.41). The increased risk of incident lung cancer for each standard deviation (6.45 mmol/mol) increase in HbA1c was more pronounced across HbA1c values of 32-42 mmol/mol (HR: 1.37, 95% CI: 1.18 to 1.59). The risk was more pronounced among participants <60 years.
CONCLUSIONS: Pre-diabetes and T2D are associated with an increased incidence of lung cancer. The increased risk of incident lung cancer is more pronounced across HbA1c values of 32-42 mmol/mol, which are currently considered normal values.