There has been a long-standing interest in gaining further insights into rare tumors whose etiology is poorly understood. At present, this project is focusing the majority of efforts on three tumors--nasopharyngeal, biliary and liver cancers. Nasopharyngeal cancer (NPC)has a very distinct geographic and ethnic distribution, occurring at high rates among ethnic Chinese from southeastern China and at much lower rates among Caucasian populations. While infection with the Epstein Barr virus (EBV) is believed to be necessary for development of the cancer, other factors, both genetic and exogenous, are also thought to be important. To investigate genetic, dietary, occupational, and behavioral factors related to the etiology of NPC, a case-control study was conducted in Taiwan. To date, our results suggest an association between risk and specific variants of the enzyme CYP2E1 and several DNA repair genes, specific patterns of HLA, and long-term cigarette smoking. High intakes of nitrosamines and nitrite during childhood and weaning also were associted with increased risks. Occupational exposures to wood dusts also appeared to affect risk; in contrast, formaldehyde exposure was not a significant risk factor. A large-scale linkage study is now under way in Taiwan to enable a careful and systematic assessment of genetic and environmental determinants of this cancer. Preliminary results appear to implicate the importance of a specific region of chromosome 14 in the etiology of the disease. During the last 25 years, biliary tract cancer incidence has increased more rapidly than any other malignancy in Shanghai. The sharply rising trend suggests a change in the prevalence of risk factors. To elucidate these factors, we recently completed fieldwork for a population-based interdisciplinary study of biliary tract cancer, the largest and most comprehensive ever. Earlier molecular analyses from tumor tissue collected in the study showed that the prevalence of mutations of several genes, including ?-catenin, p53, p16, and K-ras, varies by anatomic subsite and histology, suggesting that the molecular and causal pathways of biliary tract neoplasms may differ by anatomical subsite and histological subtype. More than 3,000 subjects were enrolled in the study. The study has a strong biochemical and molecular component with an extensive collection of biological samples, including serum, DNA, urine samples, gallstones, bile, and tissue samples. The extension collection of biological specimens, carefully collected high-quality exposure data, and the large size of the study will permit testing of a number of emergent hypothesis related to biliary tract cancer, including Helicobacter S., infection with typhoid fever and hepatitis B virus (HBV), and genetic factors. To date, a number of initial analyses have been completed, showing higher risks of biliary tract cancers associated with gallstones, a family history of gallstones, multiparity (for gallbladder cancer only), obesity, and a history of cholecystitis or pancreatitis; in contrast, tea drinking is associated with reduced risk among women. Chronic infection with HBV has also been found to be associated with a 2-fold risk of extrahepatic bile duct cancer. Although major risk factors for HCC in underdeveloped countries are well-characterized, it is still unclear why some high-risk persons develop tumors while others do not. Therefore, we are conducting studies in a high-risk population in China to examine the effects of dietary and environmental exposures, as well as genetic susceptibility on the risk of HCC. In an attempt to identify factors related to the increasing risk of HCC in the developed world, we are conducting record-linkage studies in Denmark and the U.S. These studies may help to determine whether hepatitis C virus is responsible for the increasing rates or whether other factors, such as obesity, diabetes mellitus and fatty liver disease are contributing to risk. We also also conducting similar record-linkage studies of CC in Denmark and the U.S. as the etiology of CC is largely unknown in these areas. In addition, we are analyzing age-period-cohort models of CC incidence rates in the U.S. to determine whether the increased risk is more related to a cohort effect than to a calendar period effect.
Showing the most recent 10 out of 15 publications
