Background: The role of insulin resistance as a mediator in the association of body mass index (BMI) with site-specific cancer risk has, to our knowledge, never been systematically quantified.
Methods: Altogether 510 471 individuals from six European cohorts, with a mean age of 43.1 years, were included. We used the triglyceride glucose product (TyG index) as a surrogate measure for insulin resistance. We fitted Cox models, adjusted for relevant confounders, to investigate associations of TyG index with 10 common obesity-related cancers, and quantified the proportion of the effect of BMI mediated through TyG index on the log-transformed hazard ratio (HR) scale.
Results: During a median follow-up of 17.2 years, 16 052 individuals developed obesity-related cancers. TyG index was associated with the risk of cancers of the kidney HR per one standard deviation increase 1.13, 95% confidence interval: 1.07 to 1.20], liver (1.13, 1.04 to 1.23), pancreas (1.12, 1.06 to 1.19), colon (1.07, 1.03 to 1.10) and rectum (1.09, 1.04 to 1.14). Substantial proportions of the effect of BMI were mediated by TyG index for cancers of the pancreas (42%), rectum (34%) and colon (20%); smaller proportions for kidney (15%) and liver (11%). Little or no mediation was observed for breast (postmenopausal), endometrial and ovarian cancer. Results were similar for males and females, except for pancreatic cancer where the proportions mediated were 20% and 91%, respectively.
Conclusions: The TyG index was associated with increased risk of cancers of the digestive system and substantially mediated the effect of BMI, suggesting that insulin resistance plays a promoting role in the pathogenesis of gastrointestinal cancers.
Keywords: Obesity; cancer; insulin resistance; longitudinal study; mediation analysis.
© The Author(s) 2019; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association.