筛检对肝癌死亡率影响的研究

5581名HBsAg阳性的男性随机分入周期性筛检组(A组,3712人)及对照组(B组,1869人)。A组(19155.4人年)共发生肝癌257例,B组(9785.5人年)为117例,两组的肝癌发生率分别为1342/10万与1196/10万;两组肝癌死亡分别为218与109例,肝癌死亡率分别为1138/10万与1114/10万。两组中Ⅰ期肝癌病例分别为29.6％与6.0％,差异有非常显著性意义。1、3、5年相对生存率A组为23.7％、7.0％、4.0％,B组为9.7％、4.0％、4.1％。用Poisson回归模型拟合显示,在调正年龄、初筛AFP及入列年份后,筛检对于肝癌的相对危险度为0.83,95％CI为0.68～1.03,有较弱的“保护”作用,Cox回归模型拟合结果显示当临床分期未引入模型时,筛检对于肝癌有显著的“保护”作用:危险率为0.6617,95％CI为0.5234～0.8365;而模型经调整后,危险率即接近“1”,95％CI为0.74～1.26。

Study on Effect of Screening on Mortality of Primary Cancer of Liver

A randomised controlled screening trial for primary liver cancer (PLC) was carried out among male HBsAg carriers aged 30 -69 years during the period of 1989 - 1995 in Qidong. Identified from high risk population screening using alpha-fetoprotein (AFP) assay, 5581 HBsAg carrier subjects were randomly assigned to screening group (Group A, 3712), and control group (Group B, 1869). Three hundred and seventy-three PLC cases were found during a 28941 person-year follow-up obseration, in which 257 PLC were from Group A with an incidence rate of 1341.65, and mortality rate of 1138.06 per 100000 population at risk,and 117 were from Group B with an incidence rate of 1195.64, and mortality rate of 1113.89 per 100000. The percentage of stage I PLC patients were significantly higher in Group A(29.6% )than in Group B(6. 0%). The one-, three-, and five-year relative survival rates were 23.7%,7.0%, and 4.0%in Group A, and 9. 7%, 4.0%, and 4.1% in Group B showing no difference for five-year survival between two groups. Estimation from Poisson regression model showed that the rate ratio of died of PLC in screen group was 0.83 with a95%CI of 0. 68-1.03 relative to control group. Cox regression showed group A had lower hazard ratio (0. 6617, 95% CI:0. 52 - 0. 84), but after adjusting for clinical stage and other factors, the hazard ratio covered base line 1(0.97,95%CI:0.74 -1.26). Results from our experimental study imply that screening for PLC may have benefit for early detection, early diagnosis, and early treatment, but may have no real effect, at present, on mortality of PLC for a defined population because of no real effective therapy for all patients found in screening.