Dietary components modulate the risk of hepatocellular carcinoma in cirrhotic patients.

Eighty percent of hepatocellular carcinoma (HCC) cases occur after cirrhosis from various etiologies. The association between diet and cancer is well accepted, but the links with cirrhosis progression and HCC risk have been poorly investigated. However, we hypothesized that diet could be a modifiable preventive factor for HCC. Thus, the aim of our study was to explore the relationships between dietary factors and the risk of HCC in a population of cirrhotic patients. A total of 582 cirrhotic patients were studied: 401 without HCC (controls) and 181 with HCC (cases). These patients were recruited between 2008 and 2012 for the "CiRCE" case-control study conducted in six French university hospitals. Information about the consumption of 208 food items and 23 nutrients were collected through a diet history questionnaire. Unconditional multivariate logistic regressions were performed for each residual food group and nutrients in tertiles. HCC patients were more often men, diabetic and older than controls. After adjustment, a significant positive association was found between HCC risk and carbonated beverages (OR = 2.44 [1.17-5.09] p-trend = 0.021), total cereals (OR = 1.87 [1.09-3.22] p-trend = 0.035), processed meat (OR = 1.97 [1.14-3.41] p-trend = 0.028) and sodium (OR = 2.00 [1.14-3.53] p-trend = 0.043). Conversely, the consumption of fiber (OR = 0.49 [0.28-0.86] p-trend = 0.012), vitamin E (OR = 0.52 [0.30-0.89] p-trend = 0.017), vitamin B9 (folate and folic acid) (OR = 0.56 [0.33-0.95] p-trend = 0.036), manganese (OR = 0.56 [0.32-0.97] p-trend = 0.038) and potassium (OR = 0.44 [0.25-0.76] p-trend = 0.004) were significantly lower in HCC patients compared with cirrhotic controls. Although these findings must be confirmed in prospective studies, using dietary patterns or biological parameters, they suggest that certain dietary components may modulate HCC risk in cirrhotic patients.