Association of second-hand tobacco smoking with liver cancer, and interaction with HBV infection in a Chinese population-based case–control study

Background: – Although tobacco smoking is one of the established risk factors for liver cancer, results from epidemiological studies remain inconclusive on the relationship between second-hand smoke (SHS) and liver cancer, particularly among nonsmokers. Objectives: – This study aimed to examine the association between SHS and liver cancer, and to assess its potential interaction with major risk factors such as hepatitis B virus (HBV) infection. Methods: – A population-based case–control study was conducted in Jiangsu, China, from 2003 to 2010, including 2011 newly diagnosed primary liver cancer cases and 7933 population-based controls. Data on SHS exposure at home and in the workplace, along with other major liver cancer risk factors such as alcohol consumption, were collected through self-reported questionnaires. SHS exposure was assessed as a dichotomous variable, categorizing participants as either exposed or unexposed to SHS. HBV infection status was determined by testing serum samples for serum hepatitis B virus surface antigen. Multivariable unconditional logistic regression models were used to examine the association between SHS and risk of liver cancer. Besides stratified analyses, interactions on additive and multiplicative scales were further evaluated between SHS and other risk factors for liver cancer. Results: – Exposure to SHS was associated with liver cancer, shown as adjusted odds ratios (ORs) of 1.84 (95% confidence interval [CI]: 1.56, 2.18) in the overall population and 2.11 (95% CI: 1.67, 2.65) among never smokers. Stratified analyses showed that the association between SHS and liver cancer was stronger among HBV-positive participants (OR: 2.09, 95% CI: 1.48, 2.93) compared with HBV-negative participants (OR: 1.87, 95% CI: 1.54, 2.27) (P = 0.02). Both super-additive and super-multiplicative interactions were observed between SHS and HBV infection, with relative excess risk due to interaction (RERI) of 12.56 (95% CI: 6.60, 18.53) and ratio of ORs of 1.55 (95% CI: 1.07, 2.23) in the total population, and RERI of 12.87 (95% CI: 4.93, 20.81) and ratio of ORs of 1.78 (95% CI: 1.06, 2.99) among never smokers. Conclusion: – SHS is independently associated with liver cancer, and this association is further modified by HBV infection, leading to a substantially elevated risk, particularly among nonsmokers.