Visceral and ectopic fat are more predictively associated with primary liver cancer than overall obesity from genetic sights: A Mendelian randomization study

Fei‐Qi Xu,Qing‐Yun Xu,Zhang‐Ji Zhu,Lei Jin,Tai‐Wei Ye,Cheng‐Fei Du,Zhen‐Yu Gao,Xiao‐Kun Huang,Zhe Zhang,Li‐Ming Jin,Wei‐Feng Yao

DOI: https://doi.org/10.1002/ijc.34751

2023-10-12

International Journal of Cancer

Abstract:What's new? Obesity is associated with increased risk of primary liver cancer (PLC). However, the causality of obesity in PLC is difficult to determine based on existing observational studies, and little is known about whether specific fat distribution impacts PLC risk. Here, the authors employed Mendelian randomization to investigate associations between PLC risk and genetically determined liver fat deposition and visceral fat distribution. Genetically predicted liver fat and visceral adipose tissue (VAT) distribution were more strongly linked to increased PLC risk than body mass index. VAT and especially fat deposition in the liver are promising clinical measures for predicting PLC risk. Several observational studies have reported an association between obesity and primary liver cancer (PLC), while the causality behind this association and the comparison of the risk effects of different obesity indicators on PLC remain unclear. In this study, we performed two‐sample Mendelian randomization (MR) analyses to assess the associations of genetically determined liver fat, visceral adipose tissue (VAT), and body mass index (BMI) with the risk of PLC. The summary statistics of exposures were obtained from two genome‐wide association studies (GWASs) based on the UK Biobank (UKB) imaging cohort and the Genetic Epidemiology Research on Adult Health and Aging (GERA) cohort. GWAS summary statistics for PLC were obtained from FinnGen consortium R7 release data, including 304 PLC cases and 218 488 controls. Inverse‐variance weighted (IVW) was used as the primary analysis, and a series of sensitivity analyses were performed to further verify the robustness of these findings. IVW analysis highlighted a significant association of genetically determined liver fat (OR per SD increase: 7.14; 95% CI: 5.10‐9.99; P = 2.35E‐30) and VAT (OR per SD increase: 5.70; 95% CI: 1.32‐24.72; P = .020) with PLC but not of BMI with PLC. The findings were further confirmed by a series of MR methods. No evidence of horizontal pleiotropy between these associations existed. Our study suggested that genetically determined liver fat and VAT rather than BMI were associated with an increased risk of PLC, which suggested that visceral fat distribution is more predictive of the clinical risk of PLC than common in vitro measures.

oncology