SECONDARY PREVENTION OF LIVER CANCER IN HAIMEN, CHINA

海门县原发性肝癌的二级预防

Secondary Prevention of Primary Liver Cancer in Haimen County, China Primary hepatocellular carcinoma (PHC) causes 250,000 to 1,000,000 deaths annually worldwide and 100,000 to 200,000 per year in China. Five-year survival rates for patients with PHC who are diagnosed after symptoms appear are < 5%. Surgical resection of small, asymptomatic tumors, however, has yielded five-year survival rates as high as 67%. Because other therapies have not changed survival, early detection of small (< 3 cm in diameter) tumors followed by surgical resection is the only available method of improving the outcome of these patients. Chronic hepatitis B virus (HBV) infection is associated with ~80% of liver cancers. Monitoring HBV carriers with serum alpha-fetoprotein (AFP) levels for the detection of early stage PHCs has been used among Alaskan natives and is widely practiced by clinicians in the USA. In Asia, clinicians use real time ultrasonography (US) for screening for PHC. Although US is both a more sensitive and more specific initial screening test than AFP, it is not feasible to apply US screening to large numbers of persons outside of a hospital setting. AFP as the initial screening method followed by US of persons with elevated AFPs is practical. However, there has never been a properly controlled trial of the efficacy of AFP screening alone or in combination with US. Only a controlled clinical trial which prospectively follows both a monitored and unmonitored cohort will be able to discern the ultimate efficacy of these screening modalities. Therefore, we have designed a controlled trial of AFP screening followed by US, for the early detection of PHC, to be carried out in a highly endemic area of China, Haimen County in Jiangsu Province. Two cohorts of 5000 male and 2500 female HBV carriers each, one monitored at six- month intervals, the other unmonitored (usual care) will be studied for five years. The following hypotheses will be tested: 1. The monitored will have a > 50% reduction in mortality from PHC compared with the unmonitored cohort. 2. Decreased mortality from PHC will cause reduction in all-cause mortality of the unmonitored cohort. 3. Among the monitored cohort, > 60% of the screening detected cases of PHC will have solitary resectable tumors. Among the unmonitored cohort < 10% of the PHCs will be solitary resectable lesions. 4. The predictive value positive of the screening program, AFP screening followed by US, will be > 75%. The sensitivity and specificity of the screening program will be > 80%. In addition, the following parameters of the natural history of PHC will be estimated: a. The lead time in diagnosis gained by screening. b. The length of the detectable pre-clinical phase. c. The proportion of screening detected tumors that progress to clinical cancers.