Mendelian randomization identifies genetically predicted calorie dietary preference as causal risk factors for cancer

Abstract

Background: Calorie preference refers to an individual's systematic inclination toward selecting foods based on their caloric density. The causal impact of dietary calorie preference on cancer development remains uncertain. Methods: In this study, data on dietary calorie preference were sourced from a large-scale genome-wide association study (GWAS) database of food liking, while information on 18 cancer types was obtained from Finger R9 database. Two-sample Mendelian randomization (MR) analysis was conducted to assess the causal effects of Calorie Dietary Liking on 18 types of cancers. The inverse variance weighting (IVW) method was the primary analytical approach, with significant correlations further examined using Egger regression, MR-PRESSO, and weighted median methods. Results: Our analyses revealed that a genetically predicted preference for high-calorie diets significantly raised the risk of non-Hodgkin lymphoma (NHL) (OR = 1.75, 95% CI 1.175 to 2.598), non-small cell lung cancer (NSCLC) (OR = 1.39, 95% CI 1.147 to 1.693), hepatocellular carcinoma (HCC) (OR = 2.06, 95% CI 1.097 to 3.866), and gastric cancer (GC) (OR = 1.47, 95% CI 1.087 to 1.991). Further subgroup analyses confirmed that high-calorie foods liking, particularly cheese liking, were strongly linked to an elevated HCC risk (OR = 2.43, 95% CI 1.509 to 3.899), while deep-fried food liking were associated with increased NSCLC risk (OR = 1.26, 95% CI 1.073 to 1.490). Low-calorie dietary liking showed causal association with the risk of prostate cancer (PCa) (OR = 0.85, 95% CI 0.745 to 0.965). Conclusion: This comprehensive MR analysis suggested that genetically predicted high-calorie food liking, as well as its subgroups, may be a risk factor in the development of NHL, NSCLC, HCC, and GC. A low-calorie diet may have a protective effect on the risk of PCa.