Tomio Hirohata

Nutrition and stomach cancer.

Epidemiologic evidence on the relation between nutrition and stomach cancer is reviewed. Stomach cancer shows a distinct international variation and dramatic worldwide decline. These descriptive features suggest that dietary factors are important in determining the risk of stomach cancer. The authors assessed relevant data regarding specific dietary hypotheses in the etiology of stomach cancer. A negative association with fresh vegetables and fruits is highly consistent in numerous case-control studies in different populations. Both epidemiologic and experimental data suggest that vitamins C and carotenoids lower risk of stomach cancer. Evidence is sparse and inconsistent as to protective effects of vitamin E and selenium. Epidemiologic studies have not lent, and will not provide, supportive evidence for an etiologic role of nitrate intake. High salt intake has been associated with an increased risk in many case-control studies and limited cohort studies. Taken together with animal data, it is considered that high salt intake is a risk factor for stomach cancer. Both epidemiologic and experimental data are inconclusive as to whether high-starch diets confer an increased risk. Cohort studies using quantitative dietary assessment and biologic measurement of micronutrients are needed for further understanding of etiologic roles of dietary factors in the causation of stomach cancer.

Influence of alcohol intake, cigarette smoking, and occupational status on idiopathic osteonecrosis of the femoral head.

An epidemiologic study compared 112 patients with idiopathic osteonecrosis (ON) of the femoral head having no history of systemic corticosteroid use and 168 hospital controls. Patients and controls were matched for gender, age, ethnicity, hospital, and time of initial diagnosis. The role of alcohol intake, cigarette smoking, and occupational status was assessed in relation to the development of ON. The relative risk (RR), the measure of association between ON and the risk factors, was statistically adjusted for the potential confounding effects of other factors by the conditional logistic regression model. An elevated risk for regular drinkers (RR = 7.8, p < 0.001) and a clear dose-response relationship was noted (test for trend; p < 0.001): the RRs were 3.3, 9.8, and 17.9 for current consumers of <400, 400–1000. and ≤000 ml/week of alcohol, respectively. A significantly increased risk was found for current smokers (RR = 3.9; p < 0.05). However, the cumulative effect of smoking was not evident. No increased risk was found for obesity or for heavy physical work. Regarding causation of ON, this study confirmed the consistent association with excessive alcohol intake and suggested the immediate untoward effects of smoking. The role of heavy physical work as a form of mechanical stress was not correlated with ON.

The Relationship between Aryl Hydrocarbon Hydroxylase and Polymorphisms of the CYP1A1 Gene

We examined the relationship between aryl hydrocarbon hydroxylase (AHH) and the frequency of a Msp I mutation in the 3′‐flanking region of cytochrome P450 (CYP) 1A1 (Mspl polymorphism) and another mutation in exon 7 (Ile‐Val polymorphism) in 84 healthy male subjects in Fukuoka, Japan. AHH inducibility (3‐methylcholanthrene (MC)‐induced AHH activity/non‐induced AHH activity) was correlated with the MspI polymorphism (P<0.0001) and age class (P=0.015), whereas no correlation was found for the Ile‐Val polymorphism (P=0.509). Age‐adjusted AHH inducibility (mean±SE) of the predominant homozygote (genotype A), the heterozygote (genotype B) and a homozygote rare allele (genotype C) genotypes was 4.89±0.36, 4.82±0.29 and 13.61±1.44, respectively. The genotype C showed much higher AHH inducibility than genotypes A and B (P<0.001), while no significant difference was observed between genotypes A and B. Non‐induced AHH activity was also correlated with these polymorphisms. The AHH activity of a homozygous mutant Val/Val genotype (0.076±0.010 pmol/min/106 cells) was significantly higher (P<0.05) than that of the wild‐type homozygous Ile/Ile (0.044±0.004 pmol/min/106 cells) and heterozygous Ile/Val (0.047±0.007 pmol/min/106 cells) genotypes. Our study suggests that the genotypes C and Val/Val, which are more frequent in smoking‐related lung cancer, are closely related with high AHH inducibility and high non‐induced AHH activity, respectively. Thus, the positive relationship between AHH inducibility and lung cancer is supported by our study. If our results are confirmed and the assessment of genotype becomes feasible on a population basis, identification of smokers who have genetically high susceptibility to lung cancer (genotype C or Val/Val) may become important for the prevention of lung cancer.

Epidemiology of Gallbladder Polyps: An Ultrasonographic Study of Male Self-Defense Officials in Japan

The prevalence and risk factors of gallbladder polyps diagnosed by ultrasonography were investigated in 2739 male self-defense officials who received a retirement health examination at the Self-Defense Forces Fukuoka Hospital, Japan, between October 1986 and December 1990. Excluding 38 men whose gallbladder had been removed previously, 143 men were found to have gallbladder polyps. The overall prevalence of gallbladder polyps was 5.3%. The relation between gallbladder polyps and smoking, alcohol use, body mass index, glucose tolerance, and serum lipids was examined in 137 men with stoneless polyps and 2495 normal subjects. Whereas smoking tended to be inversely associated with gallbladder polyps, none of the other lifestyle and clinical variables were related to this condition. Thus the reported risk factors of gallstones had no relation to gallbladder polyps.

Obesity and adenomatous polyps of the sigmoid colon

The relation between obesity and adenomatous polyps of the sigmoid colon was investigated in male self‐defense officials who received a retirement health examination at three hospitals of the Self‐Defense Forces in Japan between January 1991 and December 1992. Body mass index (BMI) and waist‐hip circumference ratio (WHR) were used as indices of obesity. A total of 228 adenoma cases and 1484 controls with normal sigmoidoscopy were identified in 2228 men: cases having small adenomas (<5 mm in diameter) and those with large adenomas (5 mm or greater) numbered 115 and 102, respectively. Smoking, alcohol use, physical activity, rank, and hospital were controlled for by multiple logistic regression analysis. BMI and WHR were classified into four levels using the 30th, 60th, and 90th percentiles of each distribution in the control as cut‐off points. There was a significant two‐fold elevation in the overall adenoma risk among men at the highest BMI level (≥26.95) compared with those at the lowest level (<22.48), but the risk did not linearly increase: a similar increase was also noted for large adenomas. While WHR was only weakly related to the overall adenoma risk, the risk of large adenomas progressively increased with increasing levels of WHR: odds ratio (OR) 2.9 (95% confidence interval (CI) 1.4–5.9) for the highest (≥0.958) versus lowest (<0.878) levels. BMI was not materially associated with adenoma risk after additional adjustment for WHR, but a positive association between WHR and large adenomas was independent of BMI: OR 3.4 (95%CI 1.5–7.6) for the highest versus lowest levels. These findings suggest that obesity is associated with an increased risk of colon adenomas, probably with adenoma growth.

The relationship between CYP1A1 aryl hydrocarbon hydroxylase activity and lung cancer in a Japanese population.

Because aryl hydrocarbon hydroxylase (AHH) is considered to be responsible for the activation of benzo(a)pyrene and other polyaromatic hydrocarbons in cigarette smoke to carcinogens, it is important to examine CYP1A1 (AHH) activity in the determination of susceptibility to lung cancer. We investigated AHH activity in peripheral mitogen-treated lymphocytes in 108 lung cancer patients and 95 healthy control individuals. Non-induced AHH activity was detectable in all the samples. AHH inducibility (3-methylcholanthrene-induced/non-induced AHH activity) showed a very wide interindividual variation as well as non-induced AHH activity. No significant associations were found between adjusted AHH activity and histologic type of tumor among lung cancer patients. Adjusted AHH inducibility of genotype C [geometric mean and 95% confidence interval (CI); 15.56 and 11.69-20.71] in MspI polymorphism was significantly higher than those of the other two genotypes (P = 0.0001), while no significant difference was observed between genotypes A (4.76 and 3.82-5.93) and B (5.60 and 4.57-6.86). On the other hand, non-induced AHH activity of genotype Val/Val (0.121 and 0.082-0.178 pmol/min/10(6) cells) in isoleucine-valine (Ile-Val) polymorphism was significantly higher than those of genotypes Ile/Ile (0.042 and 0.034-0.052 pmol/min/10(6) cells) and Ile/Val (0.040 and 0.030-0.053 pmol/min/10(6) cells) (P < 0.0001). Even after controlling for age, cigarettes smoked per day and season of the year, high AHH inducibility (7.0 < versus 0 < < or = 3.0: OR and 95 %CI, 12.4 and 2.88-53.4) was an independent risk factor for lung cancer. The data indicate that high AHH inducibility may strongly associate with the susceptibility to lung carcinogenesis.

Cigarette smoking, alcohol use and adenomatous polyps of the sigmoid colon.

The relationship of adenomatous polyps of the sigmoid colon with cigarette smoking and alcohol use was investigated in male self‐defense officials in Japan. In the comparison between 116 cases and 930 controls, total ethanol intake was not at all associated with the risk of adenomatous polyps, but cigarette smoking was strongly related to adenomatous polyps. After adjustment for total ethanol intake, body mass index and rank, odds ratios (and 95% confidence interval) for the categories of 0, 1–399, 400–799, and 800 or more cigarette‐years were 1.0 (referent), 2.3 (1.1–4.6), 2.9 (1.5–5.4) and 3.2 (1.6–6.5), respectively. Among five alcoholic beverages (sake, shochu, beer, whiskey including brandy, and wine), only whiskey consumption was weakly related to the risk of adenomatous polyps. Because the present findings disagree with an earlier observation on self‐defense officials, we examined the association with smoking and alcohol use separately for small (<5 mm) and large (≥5 mm) adenomas, combining data from these two studies. Cigarette smoking was more strongly associated with small adenomas while the positive association with certain alcoholic beverages were largely confined to large adenomas. These findings suggest that cigarette smoking and alcohol use may be linked with the development of adenoma at different stages of colon tumorigenesis.

Anthropometric indicators of breast cancer risk in postmenopausal women in Hawaii

As part of a case-control study of breast cancer in Hawaii, self-reported data on height, weight at various ages, breast size, shoe size, and triceps skinfold thickness were collected from 138 Japanese postmenopausal cases, 134 Caucasian postmenopausal cases, 154 Japanese neighborhood controls, and 142 Caucasian neighborhood controls. In a multiple covariance analysis, cases of both ethnic groups were slightly heavier (at all ages) and more obese (based on a weight-corrected-for-height index) than were controls, although none of the differences was statistically significant. Among the Japanese only, cases were also taller, had a greater body surface area (computed from the height and weight data), and had a larger shoe size than did the controls. The latter finding was statistically significant (p less than 0.05). Odds ratios were computed by multiple logistic regression analysis and revealed no additional relationships; however, there were suggested dose-response gradients for height, weight at age 20, and body surface area in the Japanese women and for breast size in the Caucasian women. A further analysis of risks based on changes in relative body weight between young adult life (age 20) and current age was also negative. Overall, these findings offer only weak support for an association between weight or obesity and breast cancer risk and suggest that anthropometric indices are at best very indirect indicators of true etiologic factors for breast cancer.

Risk factors for hepatocellular carcinoma among Japanese women.

To elucidate the risk factors for hepatocellular carcinoma (HCC) among women, we made a combined analysis of the data from three case-control studies conducted in high-risk areas of Japan. A total of 120 cases and 257 controls were included in the analysis. After adjustment for the study category, age, and other potential confounders, significantly increased risks were associated with chronic hepatitis-B virus infection (odds ratio [OR]=42.4, 95 percent confidence interval [CI]=11.2–160.2), a past history of blood transfusion (OR=3.7, CI=1.8–7.5), and a history of smoking (OR=2.2, CI=12–4.1). In addition, women with a history of heavy drinking experienced an elevated risk of borderline significance (OR=4.2, CI=0.9–20.4, P=0.07). When these ORs were compared with the corresponding estimates among males from the same case-control studies, no significant differences were observed between the two genders. Among the factors examined in this analysis, drinking and smoking habits—which are more common among Japanese men than women—may partly account for a large male-predominance in the incidence of HCC. Further studies are needed to clarify the roles that sex-hormones and hepatitis-C virus infection might play in the large gender difference of HCC occurrence.

Anthropometric Indicators of Breast Cancer Risk in Japanese Women in Fukuoka

This study was based on a case‐control study of breast cancer risk in Fukuoka, which was initiated to identify risk factors for the disease. We re‐analyzed the data of the study to evaluate the association of breast cancer with anthropometric variables in postmenopausal women. Data were obtained through interviewing 121 cases with historically confirmed breast cancer and an equal number of age‐matched hospital controls as well as neighborhood controls. The variables used for analysis were; height, weight at various ages, shoe size, and brassiere size and body size indices (body surface area and various body build indices) computed from the height and weight data. In a multiple covariance analysis, the results showed that cases were slightly lighter in weight and smaller for some of the indices, but the differences were not statistically significant. The odds ratios, estimated by multiple logistic regression analysis, suggested that the cases were slightly shorter in height and smaller in shoe size without any consistent dose‐response gradients. The effect of changes in weight and body size indices with aging were also assessed, but no significant findings were observed. The results reported here do not support the hypothesis of the positive association between obesity or large body size and breast cancer.