| Primary liver cancer is the fifth most common cancer and the third leading cause of cancer death worldwide. China has approximate 55% of PLC cases in the world and the southeastern area, where Haimen City is located, is among those with the highest PLC mortality rates in China. The time trend for incidence and mortality of PLC differed by regions or countries, partly depending on the change of risk factors. PLC mortality in Haimen City, where several interventions have been implemented to reduce PLC risks since 1980s, has been demonstrated decreasing since the year 1990. The time trend for PLC mortality in most recent years, however, has not been well analyzed. Regardless of the potential decreasing PLC mortality, the relatively much higher PLC risk in Haimen still suggests that the exploration of risk factors needs to be conducted both in men and women intensively. Among all the risk factors, HBV and male sex are the two predominant ones leading to a higher PLC mortality. In terms of their potential interaction, especially in additive scale, however, has not been assessed yet. Therefore, this study analyzed the time trend for PLC mortality in male subjects, conducted multi-variable analysis to find potential risk or preventive factors, and also assessed the additive interaction between HBV infection and male sex on PLC mortality. Besides HBV infection, acute hepatitis in adulthood may also bring a long-term effect on health. The authors tried to assess this effect on PLC and chronic liver diseases (CLD) as well.Data from a well-established cohort with a 16-year's follow-up in Haimen City, Jiangsu province was analyzed.The methods and results for each section were summarized as follows:1. Description of Haimen Cohort89789 subjects (60076 men and 29713 women) aged between 25 and 69 in 1992-1993 were enrolled at baseline, with an overall HBsAg prevalence of 15.8%. Until Dec.31st 2008, 6046 subjects were lost to follow-up and the annual rate was 4.6‰. A total of 8920 subjects died, including 2035 PLC death (men:1721; women:314). The crude PLC mortality in men and women were 198.44 and 72.62 per 100,000 person-years, respectively, which were even higher in older subjects.PLC and CLD were leading causes of death for HBsAg carriers, accounting for 71.20% and 61.18% of HBsAg carriers'deaths in men and women, respectively.2. Time trend of PLC mortality for male cohort,1994-2008All the male subjects who survived Jan.1st,1994 were divided into 9 birth cohorts by year of birth. The period of follow-up (1994-2008) was divided into 5 time intervals as well. PLC mortality for each birth cohort in every time interval was calculated. A standardized mortality rate (SMR) was calculated as well and then used to assess the time trend.SMR of PLC for all male subjects declined from 222.78 per 100,000 person-years during 1994-1996 to 166.50 per 100,000 person-years during 2006-2008. Percent Chang and Annual Percent Change for SMR were-20.8% and-2.5%, respectively. Time trend for PLC mortality varied in 9 birth cohorts. Men born before 1960 experienced an evident decreasing PLC mortality, while those who born after 1960 had an increasing risk. However, compared with people born ahead, PLC mortality at the same age group was still lower in those born after 1960, suggesting that the overall PLC mortality for male subjects decreased anyhow. These distinctive trends might be related to the higher prevalence of HBsAg in those born after 1960, which needs to be further explored.3. Risk and preventive factors for PLC in Haimen CohortCox proportional hazard model was used to explore potential risk or preventive factors for PLC in men and women separately, and to calculate hazard ratio (HR) and 95% confidence interval (95% CI).For men, potential risk factors included HBV infection (HR=15.97,95%CI:14.29, 17.85),peasants as occupation (HR=1.32,95%CI:1.17,1.49); cigarette smoking (compared with non-smokers, HRs for those who smoked 1-10,11-20 and more than 20 pack-years, were 1.18,1.20,1.31, respectively, with a P-trend of 0.0006 for dose-effect), previous history of acute hepatitis (compared with those who had no history, HRs for subjects had hepatitis A, hepatitis B, and unknown type of hepatitis were 13.6,2.07 and 1.97) and family history of PLC (HR=1.97,95%CI:1.68,2.32). Light alcohol consumption (<48g ethanol each time) might reduce PLC risk (HR=0.85,95%CI:0.75,0.95) while other factors including drinking water source, staple food source and regular tea consumption were not statistically significant.For women, increased PLC mortality was only related to HBV infection (HR=21.63, 95%CI:16.16,28.96) and previous history of acute hepatitis (compared with those who had no history, HRs for subjects had hepatitis A, hepatitis B, and unknown type for hepatitis were 1.60,2.96 and 2.12). All the other factors were not statistically significant. However, regular tea consumption still showed potential preventive effect on women, which reduced PLC mortality by 61.1% compared with non-drinkers.4. Additive interaction between male sex and HBV infection on PLC mortalityIn this study, an age-stratified Cox proportional hazards model was used to assesse the combined effect of sex and HBV infection on PLC mortality by calculating 3 interaction measures:the relative excess risk due to interaction (RERI), the attributable proportion of interaction (API), and the synergy index (SI). There was a greater risk difference between hepatitis B surface antigen carriers and non-carriers among men than among women. After adjustment for potential confounders, the RERI, API and SI were 33.27 (95% CI:22.54, 43.99),0.59 (95% CI:0.55,0.63), and 2.49 (95% CI:2.13,2.90), respectively, suggesting a significant synergistic effect of sex and HBV infection on PLC mortality.This joint effect was further examined according to age. The RERI decreased with increasing age, suggesting that the sex-HBsAg interaction was stronger in younger people, which may contribute to a lower sex disparity in the HBV-related risk of liver cancer in older people. In conclusion, HBV infection had a larger impact on liver cancer mortality in men than in women, which may explain at least part of the sex difference in liver cancer risk.5. Effect of HBV infection on age of death for PLC and Chronic Liver Diseases (CLD) deathsMortality rates of PLC and CLD for HBsAg carriers and non-carriers were calculated to assess the effect of HBV infection. Nonparametric rank sum test was performed to compare the median age of death for different causes by HBsAg status.All liver related mortality for Haimen cohort was 211.22 per 100,000 person-years, among which 74.1% were PLC deaths. Cirrhosis and chronic viral hepatitis were predominant types of CLD death. Male sex and HBsAg positivity both greatly increased the risk for PLC and CLD.Besides the effect on liver related mortality, HBV infection also resulted in an earlier age of death. For men, compared with non-carriers, HBsAg carriers died 5 to 6 years earlier for both PLC and CLD deaths. For women, however, HBV infection only caused an earlier age of death for PLC deaths, not for CLD deaths.No difference in age of death between men and women of same HBsAg status was found (P>0.05). Therefore, the sexual difference in age of death might due to the difference of HBsAg prevalence in men and women.Moreover, male HBsAg carriers and non-carriers both showed the similar pattern on age of death:chronic viral hepatitis |
PDF Full Text Request