PD1.01 Associations of Variability in Metabolic Parameters with Lung Cancer in the General Population

Variability of metabolic parameters, such as blood cholesterol, fasting blood glucose, blood pressure and body weight can affect outcomes in health. Previous studies have shown that components of metabolic syndrome can lead to increased risks for several types of cancers such as breast and colon cancer, but not many studies have shown such effects on lung cancer. This study aimed to investigate whether variability in metabolic parameters can lead to increased risk of lung cancer in the general population. Nationally representative data from the Korean National Health Insurance System was used, and a total of 8,011,209 who were not diagnosed with lung cancer and underwent over three health examinations from 2005 to 2012 were followed until the end of 2015. Variability was measured in fasting blood glucose, total cholesterol, systolic blood pressure and body weight by using the coefficient of variation, standard deviation, variability independent of the mean and average real variability, and variability was assessed by quartiles. Subjects were also classified according to number of high-variability components (for example, a score of 4 would indicate high variability in the 4 mentioned metabolic parameters). Cox proportional hazards models were used adjusting for age, sex, smoking, regular exercise, alcohol and income. There were 44,982 events of lung cancer occurrences. High variability in each metabolic parameter was associated with higher risks for lung cancer. For blood glucose, hazard ratios and 95% confidence intervals in the highest quartile was 1.07 (1.04-1.10); for systolic blood pressure, 1.08 (1.05-1.10); for body weight, 1.04 (1.01-1.07), and for total cholesterol, 1.11 (1.08-1.14). Furthermore, the risk of lung cancer increased significantly with number of high-variability metabolic components. When comparing scores of 3 and 4 versus 0 in multivariable-adjusted model, hazard ratio for lung cancer occurrence was 1.18 (95% CI 1.14-1.22). High variability in metabolic parameters, especially in total cholesterol levels, was shown to be an independent predictor of increased risk for lung cancer. Furthermore, increasing number of high-variability parameters was found to increase risk of lung cancer in a stepwise manner.