The Effect of Coffee Consumption on the Development of Hepatocellular Carcinoma in Hepatitis B Virus Endemic Area

Eun Sun Jang; Sook-Hyang Jeong; Sang Hyub Lee; Sung Ho Hwang; So Yeon Ahn; Jaebong Lee; Young Soo Park; Jin Hyoek Hwang; Jin-Wook Kim; Nayoung Kim; Dong Ho Lee; Hyun Young Kim


Liver International. 2013;33(7):1092-1099. 

In This Article


Recently, coffee has become one of the most consumed beverages. People drink approximately 2.25 billion cups of coffee everyday worldwide and Korea is a high coffee consuming country, importing more than 100 000 tonnes of coffee a year. Thus, the health effects of coffee consumption get more attention from the public. In the present case–control study, we observed the effect of high coffee consumption on the risk of HCC among Korean HCEs and CLD patients. Adjusting for age, gender, BMI, smoking and alcohol drinking amounts, high coffee consumption reduced HCC risk significantly among HCEs and non-HBV related CLD patients. However, the protective effect of coffee consumption was not apparent in CHB patients.

Overall, the risk of HCC decreased with a high lifetime coffee consumption (≥20 000 cups) by 43% in HCEs and by 45% in chronic liver disease patients respectively. These findings were consistent with the previous studies regarding the liver cancer risk of coffee drinkers [18, 19, 23, 24]. According to previous reports, the mean risk reduction rate by high coffee consumption was about 31–55%.[9,25,26] However, most studies have been performed in Europe[13,19] where hepatitis C is the predominant form of chronic viral hepatitis. Although several large prospective cohort studies[14,18,24,27] from Japan also documented that coffee drinking was associated with reduced risk of HCC, the population consisted with more chronic hepatitis C patients rather than CHB.

Whether the beneficial effect of coffee on the HCC risk may be different according to underlying aetiology of liver disease, especially hepatitis virus infection, has not been reported in previous studies. Though both HBV and HCV infections cause chronic liver injury and subsequent hepatic fibrosis, the carcinogenic mechanisms are distinct. During the progression of liver damage, HCV augments oxidative stress and steatosis in hepatocytes.[28] In contrast, HBV can directly transform hepatocytes via integration into the host genome.[29] Because coffee constituents can prevent oxidative liver damage,[30] the antitumour effect of coffee might vary depending upon the oxidative stress during carcinogenesis. According to our subgroup multivariable analyses, high coffee consumption did not lower the HCC risk among CHB patients independently, but did significantly in other chronic liver disease patients ( Table 3 ). The HCC risk in CHB patients was more strongly affected by alcohol drinking, smoking, hepatitis B envelope antigen status and a high viral load, already well-described high risk factors of HCC.[2,31] These findings suggest that the protective effect of coffee drinking could not surpass the strong direct carcinogenic effect of HBV infection, and might be related to an anti-oxidant effect.

Although many epidemiological studies[14,15,17,24] have reported an inverse association between coffee consumption status at the time of the study and HCC risk, a large proportion of HCC patients (60%) had quit coffee drinking after the diagnosis of the cancer in this study. Therefore, we calculated lifetime coffee consumption amounts in current or past coffee drinkers based on the mean daily amount and the duration of coffee drinking. As a result, we found that a high 'lifetime' coffee consumption had a strong protective effect against HCC as well as a high daily coffee consumption.[14,15,17,24]

Although multivariable analyses were performed to adjust many confounders on the HCC risk in this study, the risk of HCC in HCE and CLD controls could not be compared directly in single statistical model. As person who infected with hepatitis virus was thought to be a chronic liver disease patient, no subjects in HCE group had hepatitis virus infection which is the most important risk factor of HCC. Moreover, the diagnosis of CLD or HCC might affect the subject's personal habits of alcohol drinking or smoking, similarly with coffee consumption. Therefore, we separated the two control groups and performed analyses with lifetime alcohol, smoking and coffee consumption amounts in each controls.

As other case–control studies had, this study had a possibility of recall bias,[32] because we collected data based on the participants' response to the questionnaire. Moreover, some subjects at the high risk of HCC such as heavy drinker or smoker might be excluded from this study as a significant proportion (about 45%) of eligible subjects had refused to respond to our questionnaire survey. However, almost all participants provided complete responses to the questionnaire in our study and the information about well-known risk factors, such as smoking or alcohol, showed effects on HCC risk compatible with previous reports. Moreover, high coffee consumption showed a strong negative association with HCC risk although it was related to high alcohol drinking and smoking. Also, we have controlled for possible rumination bias[33] in healthy controls by using CLD controls that share similar behavioural risk factors with HCC, such as drinking and smoking. Additionally, serological tests about HBV or HCV infection had been performed in all subjects. The HBV DNA or HBeAg status was tested in almost all of the HBV related liver disease patients and the effect of antiviral therapy was considered in this study. However, the longitudinal effect of coffee drinking on the HBV replication in chronic liver disease patients could not be demonstrated as the results of HBeAg and HBV DNA level were not taken when subjects started drinking coffee because of the cross-sectional design.

In conclusion, high coffee consumption was significantly associated with a reduced risk of HCC in subjects with and without chronic liver disease. However, the protective effect of high coffee consumption was not proven in CHB patients. Further studies are needed to clarify the protective mechanism of specific ingredients in coffee and the possible interaction between coffee and underlying viral aetiologies in the HCC development among chronic liver disease patients.