- (1) International Agency For Research On Cancer, grid.17703.32
- (2) Emory University, grid.189967.8
- (3) German Institute of Human Nutrition, grid.418213.d
- (4) National and Kapodistrian University of Athens, grid.5216.0
- (5) Harvard University, grid.38142.3c
- (6) Academy of Athens, grid.417593.d
- (7) Aarhus University, grid.7048.b, AU
- (8) Danish Cancer Society, grid.417390.8
- (9) Institut Gustave Roussy, grid.14925.3b
- (10) German Cancer Research Center, grid.7497.d
- (11) Hellenic Health Foundation, grid.424637.0
- (12) Istituto per lo Studio e la Prevenzione Oncologica, grid.417623.5
- (13) University of Naples Federico II, grid.4691.a
- (14) Fondazione IRCCS Istituto Nazionale dei Tumori, grid.417893.0
- (15) Department of Epidemiology, Local Health Unit To3, Turin, Italy.
- (16) University of Turin, grid.7605.4
- (17) Cancer Registy and Histopathology Unit, "Civic - M.P. Arezzo" Hospital, ASP, Ragusa, Italy.
- (18) National Institute for Public Health and the Environment, grid.31147.30
- (19) University of Malaya, grid.10347.31
- (20) University Medical Center Utrecht, grid.7692.a
- (21) Imperial College London, grid.7445.2
- (22) University of Oslo, grid.5510.1
- (23) The Arctic University of Norway, grid.10919.30
- (24) Cancer Registry of Norway, grid.418941.1
- (25) Folkhälsans Forskningscentrum, grid.428673.c
- (26) Karolinska Institute, grid.4714.6
- (27) Instituto de Salud Pública de Navarra, grid.419126.9
- (28) CIBER Epidemiology and Public Health CIBERESP, Madrid, Spain.
- (29) Catalan Institute of Oncology, grid.418701.b
- (30) Instituto Murciano de Investigación Biosanitaria, grid.452553.0
- (31) Public Health Direction and Biodonostia Research Institute and CIBERESP, Basque Regional Health Department, San Sebastian, Spain.
- (32) Public Health Directorate, Asturias, Spain.
- (33) Umeå University, grid.12650.30
- (34) Lund University, grid.4514.4
- (35) Skåne University Hospital, grid.411843.b
- (36) University of Cambridge, grid.5335.0
- (37) University of Oxford, grid.4991.5
- (38) MRC Epidemiology Unit, grid.415056.3
The role of amount and type of dietary fat consumption in the etiology of hepatocellular carcinoma (HCC) is poorly understood, despite suggestive biological plausibility. The associations of total fat, fat subtypes and fat sources with HCC incidence were investigated in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, which includes 191 incident HCC cases diagnosed between 1992 and 2010. Diet was assessed by country-specific, validated dietary questionnaires. A single 24-hr diet recall from a cohort subsample was used for measurement error calibration. Hazard ratios (HR) and 95% confidence intervals (95% CI) were estimated from Cox proportional hazard models. Hepatitis B and C viruses (HBV/HCV) status and biomarkers of liver function were assessed separately in a nested case-control subset with available blood samples (HCC = 122). In multivariable calibrated models, there was a statistically significant inverse association between total fat intake and risk of HCC (per 10 g/day, HR = 0.80, 95% CI: 0.65-0.99), which was mainly driven by monounsaturated fats (per 5 g/day, HR = 0.71, 95% CI: 0.55-0.92) rather than polyunsaturated fats (per 5 g/day, HR = 0.92, 95% CI: 0.68-1.25). There was no association between saturated fats (HR = 1.08, 95% CI: 0.88-1.34) and HCC risk. The ratio of polyunsaturated/monounsaturated fats to saturated fats was not significantly associated with HCC risk (per 0.2 point, HR = 0.86, 95% CI: 0.73-1.01). Restriction of analyses to HBV/HCV free participants or adjustment for liver function did not substantially alter the findings. In this large prospective European cohort, higher consumption of monounsaturated fats is associated with lower HCC risk.