Glycemic variability and subsequent malignancies among the population without diabetes.

Background: Glycemic variability has been suggested to be related to some unfavorable outcomes, but malignancy development has not been evaluated. The aim of this study was to evaluate the association of glycemic variability with malignancy development among the population without diabetes.

Methods: We conducted a retrospective cohort study at a large teaching hospital in Tokyo, Japan, from 2005 to 2016. We included all participants without diabetes who underwent voluntary health check-ups. Our outcome was the development of any malignancy. As a measure of glycemic variability, we calculated the quotient of CV in HbA1c and categorized subjects into quartile groups. A Cox proportional hazard model was applied, adjusting for patient demographics and social and family histories.

Results: A total of 42,731 participants were included in this study; the mean age was 53.8 and 48.3% were male. During the median follow up of 2639 (interquartile range (IQR):1787-3662) days, 2435 participants (5.7%) developed malignancies. Participants who had larger glycemic variability (CV in HbA1c; hazard ratio (HR) 1.15, 95%confidence interval (CI):1.02-0.31 for the second quartile group; HR 2.20, 95%CI:1.95-2.48 for the third quartile group, HR 4.66, 95%CI:4.16-5.21 for the fourth quartile group, compared to first quartile group) had a significantly higher risk of malignancies.

Conclusion: We found an association between large glycemic variability and a high risk of future malignancies in a dose-dependent manner among people without diabetes. This finding suggests that maintaining a constant level of glucose may have favorable effects on cancer prevention in people without diabetes.