Between 1985 and 1991, approximately 30,000 persons in Linxian, China participated in two nutritional intervention trials that tested whether various mineral and vitamin supplements would decrease the incidence of esophageal and gastric cancer. In 1996, the mortality and morbidity of these participants was assessed by interview, physical examination, and some endoscopic examinations. During this first five years post intervention there were 1,388 new incident esophageal and gastric cancers, and 1,158 new strokes. Analyses of these data show: 1) the benefits on overall mortality found in the group who received selenium, beta-carotene, and vitamin E during the course of the General Population trial have persisted five years after the end of intervention; 2) the small but statistically insignificant benefit of a multivitamin on esophageal cancer in the Dysplasia Trial has increased ; 3) there are statistically significant benefits on cancer mortality for both the multivitamin and the selenium, beta-carotene, vitamin E supplement for participants less than 55 years old. We have also conducted several studies examining the relationship between serum levels of minerals and vitamins before the supplements were given, and the subsequent risk of developing squamous esophageal, gastric cardia, and gastric non-cardia cancers. The study that examined the relationship of selenium and 1100 incident cases of cancer is one of the largest prospective studies of serum selenium and cancer risk and has more site-specific cancers than any previous study. We found highly significant inverse associations of serum selenium levels with the incidence of both esophageal and gastric cardia cancers over 5.25 years of follow-up. Individuals in the highest quartile of selenium developed these cancers at approximately half the rate as individuals in the lowest quartile. For both cancers there was a dose-response relationship. Despite probable differences between exposure-cancer relationships in the United States and in China, the importance of selenium in common pathways involving tissue damage prevention and repair renders these findings from Linxian relevant to the U.S. population. Consequently we have recommended that randomized preventive trials for selenium currently in the planning stages in the United States include populations at high risk for squamous cell carcinoma of the esophagus and adenocarcinoma of the gastric cardia. In ongoing analyses of other serum vitamins and metabolites we have found :1) significant inverse associations of both serum cysteine and serum vitamin E levels with the incidence of esophageal and gastric cardia cancers 2) significant inverse association of serum riboflavin with esophageal cancer; and 3) no association of beta-carotene with either of these cancers. Results from serum measurements on B-vitamins show that the Linxian subjects have markedly high levels of homocysteine (homocysteine is thought to be a cause of stroke and heart disease) and low levels of many B-vitamins : the average homocysteine level in Linxian is nearly twice the level found in the United States;75% of the population was deficient in vitamin B12. B vitamin deficiencies of the degree we find in Linxian are typically associated with debilitating central and peripheral nervous system deficits, and with profound disruption of hematologic maturation. In the past year we completed the field work for a study that measures nervous system function and the status of peripheral blood in 1,000 subjects. This study also will enable us to validate and further develop the nutrition questionnaire we use to assess dietary adequacy in our subjects. In addition to the intensive examination of 1,000 subjects, we have also completed a study that collected additional biological samples on all 23,000 living participants. This will allow allow us to measure vitamin C, and to preserve white blood cells as a source of DNA for future genetic studies. We are investigating the relationship of infectious diseases and fungal contaminants of grain to esophageal and stomach cancers. In a study of the relationship of H. pylori infection and the subsequent risk of gastric cardia and gastric non-cardia cancer, we found H. pylori seropositivity was associated with an approximately two fold risk for both gastric cardia cancer and gastric non-cardia cancer. We have found that serologic evidence of infection with human papilloma virus type 16 and 73 is associated with increased rates of squamous esophageal cancer and adenomatous cardia stomach cancer, but not with adenomatous non-cardia stomach cancer. This is the first study to examine HPV73 and esophageal cancer. We have found no relation of Epstein-Barr virus to gastric cancer. Ecologic studies of esophageal squamous cell carcinoma have reported an association with consumption of corn contaminated with the mold Fusarium verticillioides. This mold produces a family of toxins referred to as fumonisins which disrupt sphingolipid metabolism. On the bases of animal experiments it had been previously suggested that the ratio of sphinganine to sphingosine is a good marker of fumonisin exposures. Our study is the first prospective study to assess the relationship between sphingolipid levels as a biomarker of fumonisin exposure, and cancer incidence in humans. We found no significant association between fumonisin exposure and risk of esophageal squamous cell carcinoma. Gastric cancer is the second most frequent cause of death from cancer in the world and the leading cause of death from cancer in China, In September 1995, we launched a randomized multi-intervention trial to inhibit the progression of precancerous gastric lesions in Linqu County, Shandong Province, an area of China with one of the world's highest rates of gastric cancer. Treatment compliance was measured by pill counts and quarterly serum concentrations of vitamin C, vitamin E and S-allyl cysteine, In 1999, toxicity information was collected from each trial participant to evaluate treatment-related side-effects during the trial, Compliance rates were 93% and 92.9% for 39 months of treatment with the vitamins/mineral and garlic preparation, respectively. No significant differences in side-effects were observed between the placebo treatment group and the vitamins/mineral and garlic preparation treatment groups during the 39-month trial period. We have done several studies on morbidity following treatment of malignancies. We found that females who received cranial radiotherapy for acute lymphoblastic leukemia around the time of menarche had increased risk of subsequent fertility problems. In a study of chemotherapy for solid tumors, we found that the risk of bacteremia increased in parallel with treatment intensity.
