The Epidemiology and Prevention Division has planned and conducted independent and collaborative studies on cancer etiology and prevention, with a special focus on dietary, environmental and genetic factors. Several epidemiological projects are currently in progress.

Diet has been implicated in the etiology of cancer and in the unique patterns of cancer incidence in Japan. However, the epidemiological evidence for this contention has been limited. The division therefore initiated a cohort study, the Japan Public Health Center-based Prospective Study (JPHC Study), in 1990, in collaboration with 11 public health centers and other institutes, in which approximately 140,000 individuals from 11 areas are scheduled to be followed up for 20 years. A total of 13,045 deaths and 11,230 cancers had been documented as of 31 October, 2006. Lifestyle factors that were assessed in the baseline questionnaire were examined in relation to the subsequent risk of incidence of cancer and other lifestyle-related diseases. Subjects with a history of diabetes mellitus (DM) had a higher cancer risk than those without (relative risk (RR) =1.3 for men and 1.2 for women) ⁽¹⁸⁴⁾ and subjects with a family history of lung cancer had an approximately 2-fold higher lung cancer risk than those without a family history ⁽¹⁸⁵⁾. A 2-fold decrease in gastric cancer mortality in X-ray-screened versus unscreened subjects was observed ⁽¹⁸⁶⁾. Body mass index and height were not significantly associated with the risk of prostate cancer ⁽¹⁸⁷⁾. Neither the frequency of bowel movements nor an unusual stool forms, such as diarrhea or hard stools, was associated with the incidence of colorectal cancer ⁽¹⁸⁸⁾. No statistically significant association between dietary fiber intake and colorectal cancer was found, but the lowest subtertile in the lowest quintile was associated with an increased risk in women (RR=2.3) ⁽¹⁸⁹⁾. Two nested case-control studies using cryopreserved blood samples were performed. The group with values in the highest quartile of plasma C-reactive protein showed a significantly increased incidence of colorectal cancer as compared with that with values in the lowest quartile (RR=1.6), especially of intramucosal colorectal cancer ⁽¹⁹⁰⁾. The RR of gastric cancer associated with H. pylori infection was 5.1. This risk doubled in the additional presence of CagA-positive or atrophic gastritis as measured by the serum pepsinogen levels ⁽¹⁹¹⁾. In addition to cancer, some results on chronic diseases have also been obtained. As compared with a modest fish intake of once a week or 20 g/d, a higher intake was associated with a substantially reduced risk of coronary heart disease (CHD), primarily nonfatal cardiac events ⁽¹⁹²⁾. Current smokers had almost a 3-fold increased risk of CHD as compared to never-smokers ⁽¹⁹³⁾. Although the serum total cholesterol level was significantly related to the risk of CHD, dietary egg consumption was not associated with the risk ⁽¹⁹⁴⁾. There was a U-shaped association between alcohol consumption and subsequent suicide ⁽¹⁹⁵⁾. Our data showed that even for RRs, the extrapolation of estimates for a sample from a well-defined population to the general population may not be possible ⁽¹⁹⁶⁾. Despite changes in the intake levels of many nutrients, the validity of our FFQ using rank correlation by nutrient intake was not influenced by revision of the nutrient database in Japan ⁽¹⁹⁷⁾.

To develop an evidence-based cancer prevention strategy in terms of lifestyle intervention that is suitable for the Japanese population, a systematic literature review project and some interventional studies are in progress. The relation between tobacco smoking and the cancer risk of each of colorectal ⁽¹⁹⁸⁾, lung ⁽¹⁹⁹⁾, breast ⁽²⁰⁰⁾, stomach ⁽²⁰¹⁾ and liver ⁽²⁰²⁾ cancer was evaluated. Evidence for alcohol drinking and colorectal cancer risk was also reviewed ⁽²⁰³⁾. A population-based survey showed that awareness of the attributable fraction of cancer causes tends to be dominated by cancer-causing infections, occupational exposure, air pollution and food additives rather than major lifestyle factors, such as the diet ⁽²⁰⁴⁾. A randomized controlled trial to assess the effectiveness of vitamin C supplementation for individuals with chronic atrophic gastritis to prevent gastric cancer in the high-risk area of Hiraka, Akita prefecture, has been completed and the data are under analysis. The data suggest that while vitamin C supplementation significantly reduced the frequency of common cold, it had no apparent effect on the duration or severity of cold ⁽²⁰⁵⁾. A community-based randomized cross-over trial designed to develop an effective dietary modification tool and system was held in the same area in 1998-2000, and the follow-up survey after the intervention is in progress. The intervention of moderate-intensity dietary counseling yielded significant dietary changes, which resulted in a significant decrease of the mean systolic blood pressure ⁽²⁰⁶⁾. The low-sodium, high-vitamin C and high-carotene diet was maintained for well over 4 years after the termination of the intervention sessions ⁽²⁰⁷⁾.

The ethnic differences in the incidence of cancer suggest an interaction between environmental and genetic factors. Several epidemiologic studies in Brazil, a multi-ethnic nation with 1.2 million people of Japanese ancestry, are in progress. Case-control studies for breast cancer and colonic adenoma are also in progress.

A project to evaluate risk of endocrine-disrupting chemicals (EDC) on human health is also in progress. A cross-sectional study among Japanese women consulting doctors for infertility suggests that those who consume fish frequently tend to accumulate organochlorines in their bodies ⁽²⁰⁸⁾. A cross-sectional study among male workers in China showed a modest but significant reduction of serum free testosterone in workers with higher levels of urinary mono-n-butyl phthalate and mono-2-ethylhexyl phthalate as compared with the levels in non-exposed workers ⁽²⁰⁹⁾.

A cross-sectional study among newly diagnosed colonic adenoma cases from the examinees of a colorectal cancer screening program suggested that a higher serum triglyceride level, which may be modified by smoking, may be related to a larger number of colonic adenomas ⁽²¹⁰⁾. The adductome approach to detect DNA damage in humans was developed ⁽²¹¹⁾.