Cancer Epidemiology and Large Randomised Trials
Previous and current research:
One of the major research interest of our group is the assessment of some of the widely practicable treatments for cancer, by obtaining large-scale randomised evidence either through overviews of previous trials or by conducting large randomised trial, involving many hundred hospitals. A number of large cancer trials are under way in China, including one with 7 days of portal-vein infusion with 5-FU among 11,000 colorectal cancer patients and one with ovarian ablation among 3500 early breast cancer patients.
Another main research area is large epidemiological studies in China assessing the health effects of smoking, alcohol drinking, tea drinking, indoor air pollution, blood pressure, lung function and blood lipids on cancer and various other chronic diseases. These include a large nationwide retrospective study of one million Chinese deaths in 1986-88 and the nationwide prospective study of 250,000 Chinese adults that began in 1991. These studies indicated that although the pattern of disease produced by tobacco is very different from that in countries such as the United States and the United Kingdom (the chief risks produced in Chinese smokers being death from cancer, respiratory disease rather than vascular), the overall magnitude of the hazard is already substantial even at an early stage of tobacco epidemic.
Future projects:
All patients entered into two large trials will be followed up for at least five years for the assessment of long-term survival. In colorectal cancer trial with portal-vein infusion, 1000 pairs of normal and cancer tissue samples have been collected. This will provide a unique opportunity to evaluate the value of the known molecular features as prognostic markers (such as p53, K-ras and HNPCC mutation, etc). We intend to examine if, in China where the incidence of colorectal cancer is still very low, the frequencies of those mutations are similar to that reported in Western populations. We also intend to explore the relationship between certain types of mutation and long-term survival.
On epidemiological side, further detailed analyses will be undertaken to examine the relationship of smoking, alcohol drinking, indoor air-pollution and various other factors with total and cause specific mortality. Deaths in the large prospective study will be actively monitored for the next 5, 10 and 15 years to assess the evolution of tobacco epidemic.
