Genetic risk, metabolic syndrome, and gastrointestinal cancer risk: A prospective cohort study

Abstract

Background: Gastrointestinal (GI) cancer risk has been associated with metabolic syndrome (MetS), a surrogate indicator for unhealthy lifestyles, and a number of genetic loci, but the combined effect of MetS and genetic variants on GI cancer risk is uncertain.

Methods: We included 430,036 participants with available MetS and genotype data from UK Biobank. During the follow-up time, 5494 incident GI cancer cases, including esophageal cancer, gastric cancer, and colorectal cancer, were identified. We created a GI polygenic risk score (GI-PRS) for overall GI cancer derived from three site-specific cancer PRSs. Cox proportional hazards regression was used to estimate the associations of MetS and GI-PRS with the risk of GI cancer.

Results: MetS was significantly associated with 28% increment in GI cancer risk (hazard ratio [HR]_{MetS vs. non-MetS} : 1.28, 95% confidence interval [CI]: 1.21-1.35, p < 0.0001), whereas a high GI-PRS (top quintile) was associated with 2.28-fold increase in risk (HR_{high vs. low} : 2.28, 95% CI: 2.09-2.49, p < 0.0001). Compared with participants without MetS and at low genetic risk (bottom quintile of GI-PRS), those with MetS and at high genetic risk had 2.75-fold increase in GI cancer risk (HR: 2.75, 95% CI: 2.43-3.12, p < 0.0001). Additionally, MetS in comparison with no MetS had 1.49‰, 2.75‰, and 3.37‰ absolute risk increases in 5 years among participants at low, intermediate (quintiles 2-4 of GI-PRS) and high genetic risk, respectively, representing the number of subjects diagnosed as MetS causing a new GI cancer case in 5 years were 669, 364, and 296, respectively.

Conclusions: Metabolic and genetic factors may jointly contribute to GI cancer risk and may serve as predictors by quantitative measurements to identify high-risk populations of GI cancer for precise prevention.

Keywords: gastrointestinal cancer; genetic risk; metabolic syndrome; polygenic risk score.

FIGURE 1

The relationship of MetS and PRS with GI cancer risk. (A) Participants were divided into six groups based on the number of MetS components they had and HR of each group was compared to participants having no MetS components. (B) Standardized rates of GI cancer events in non‐MetS and MetS groups. (C) Participants were grouped based on PRS quintiles and HR of each group was compared to those at the lowest quintile. (D) Standardized rates of GI cancer events in those at low (lowest quintile), intermediate (quintiles 2–4), and high (top quintile) genetic risk. GI, gastrointestinal; HR, hazard ratio; MetS, metabolic syndrome; PRS, polygenic risk score.

FIGURE 2

The joint effect of metabolic syndrome and genetic categories on gastrointestinal cancer risk. Models adjusted for age group, gender, qualification, Townsend Deprivation Index, family history of cancer, physical activity, smoking status, alcohol consumption, fruit intake, vegetable consumption, red and processed meat consumption, regular aspirin or ibuprofen use and the top 10 genetic principal components.