Several studies have investigated the possible relationship between fertility drugs and the risk of breast cancer. To provide further information on this issue, we analysed data from a case control study, conducted in Northern Italy between 1983 and 1991. Trained interviewers identified and questioned 3415 cases (women aged 23-74 years with histologically confirmed breast cancer) and 2916 controls (women aged 21-74 years admitted to the same hospitals for diseases other than malignant, hormonal or gynaecological conditions). Fifty (1.5%) cases and 53 (1.8%) controls reported any history of infertility; the corresponding multivariate odds ratios (OR) of breast cancer was 0.8 [95% confidence interval (CI) 0.5-1.1]. Sixteen (0.5%) cases and 11 (0.4%) controls reported ever using fertility drugs; the corresponding OR was 1.2 (95% CI 0.5-2.6). Allowance for potential confounding factors did not markedly modify these estimates. In conclusion, this study provides reassuring evidence on the absence of an association between fertility drug treatment and breast cancer risk.

Increased incidence of adenocarcinoma of the gastric cardia has been reported over the last few decades from several areas of North America and Europe (1–2). A mortality study from the Swedish Cancer Registry, however, suggested that the observed upward trends can be, partly or largely, accounted for by changed accuracy of registration within gastric subsites (3). We considered, therefore, trends in incidence rates for various gastric subsites in the Cancer Registry of the Swiss Canton of Vaud (covering approximately 600 000 inhabitants in 1990 from the French-speaking part of Switzerland) over the period from 1976 through 1997. In this area, uniform criteria of classification have been adopted, and traditional attention has been focused on careful endoscopic and histopathologic examination of gastric lesions (4–6). Table 1 gives average age-adjusted (on the world standard population) incidence rates for various subsites of gastric cancer during three separate calendar periods. In both sexes, no appreciable change in incidence of adenocarcinoma of the gastric cardia was observed (3.1 cases per 100 000 males, and 0.5 cases per 100 000 females from 1976 through 1979 versus 3.2 and 0.1, respectively, from 1995 through 1997), while appreciable downward trends were observed for distal and other or unspecified gastric cancer sites. These data, from a carefully surveyed European population, therefore do not support the existence of a systematic and major rise in incidence of cardiac adenocarcinomas (3,7), confirming that—in proportional terms—the cancers of the gastric cardia have become

The study provides reassuring information on a large cohort of a unique geothermal power plant since no significant excess mortality for asbestos related cancers was observed. Furthermore, it represents a useful application and validation of a comprehensive job-exposure matrix for the electric industry in Europe.

Evidence of a protective role of aspirin on the risk of colorectal and other common cancers has been building up since the end of the 1980s. There are now more than 15 epidemiological (case-control and cohort) studies indicating that long-term use of aspirin is associated with a reduced risk of colorectal cancer. The overall relative risk (RR) estimate for regular aspirin users was 0.71 (95% confidence interval (CI) 0.66-0.77) from case-control studies, and 0.84 (95% CI 0.72-0.98) from cohort studies. A recent meta-analysis reported a RR of breast cancer for aspirin use of 0.70 (95% CI 0.61-0.81) in case-control studies, and of 0.79 (95% CI 0.59-1.06) in cohort studies. Furthermore, various epidemiological studies have suggested that aspirin use might have a favourable effect on ovarian cancer as well: the overall RR estimate was 0.82 (95% CI 0.69-0.99), although the evidence is too limited to permit firm conclusions. Data are more scanty, though in the same direction, for other neoplasms, including in particular stomach and oesophageal cancer.

The fall in cancer mortality observed over the last decade in Italy is attributable to a decline in lung and other tobacco-related neoplasms in males, together with a persistent fall in stomach and uterine (cervical) cancer. In women, there were also recent falls in intestinal and breast cancer rates, and declines in both sexes in rarer neoplasms influenced by therapeutic advancements.

The incidence of most cancers rises with the fourth to fifth power of age, and that of breast cancer rises steeply up to age of menopause and continues to rise, though more slowly, thereafter. Contralateral breast cancer, however, seems to have a completely different age incidence pattern. According to the data of the Connecticut Cancer Registry, contralateral breast cancer rate was around 1% per year in women below age 45, and decreased subsequently to approach a constant rate of approximately 0.7% per year.1 In the Swedish Family Cancer Database, including 2,529 contralateral breast cancers, the rate was 0.8–0.9% per year between age 30 and 50, and declined thereafter to approximately 0.4% per year.2 This pattern of risk is compatible with the occurrence of a single mutational event in a population of susceptible women.3, 4 To provide further information on the issue, we have examined the epidemiology of contralateral breast cancer in the Swiss Canton of Vaud. For the present analysis we used the Vaud Cancer Registry dataset, which includes incident cases of malignant neoplasms in the Canton (around 602,000 inhabitants in 1990).5 Population-based incidence data have been available since 1974. After exclusion of 166 breast cancer cases detected either at autopsy, or at death, or by death certificate alone, and of synchronous breast cancers (i.e., within 2months after the first primary, n = 242), the present series comprised 7,734 women with a first breast cancer diagnosed between 1974 and 1998 (rate of histological confirmation: 96.8 percent). These women were followed-up to the end of 1998 for the occurrence of a second contralateral primary neoplasm, emigration, or death, for a total of 49,143 person-years at risk. Among these women, 214 had subsequent contralateral breast primary. Figure 1a gives the age distribution of breast cancer in the general population, based on 7,520 first single neoplasms and no subsequent contralateral, showing the well known steep rise in incidence below age 50, and the leveling of the slope of age thereafter. Age-specific incidence of (a) first breast cancer in the general population (n = 7,520; rates in quinquennial age groups per 100,000 women); (b) contralateral breast cancer in 214 women with first breast cancer (rates in decennial age groups per 100,000 women and (c) first breast cancer in the general population in 214 women who presented with contralateral breast cancer (rates in decennial age groups per 100,000 women). Swiss Canton of Vaud, 1974–98. Figure 1b gives the incidence of contralateral breast cancer, based on 214 neoplasms. The peak rate of 800/100,000 women was reached at age 30 to 39, the rate declined and reached 500/100,000 between age 45 and 69, and further declined thereafter to 300–400/100,000, i.e. a rate similar to that of first primary neoplasms above age 70. Figure 1c gives the age distribution of rates in the general population of first breast neoplasms in the 214 women with second ones. The rate of 6/100,000 women was reached at age 40 to 49, the incidence rates tended to level off until age 70, and to decline thereafter. The present data confirm that the age incidence curve of contralateral breast cancer is substantially different from that of breast cancer in the general population, with a peak at younger age (30–39 years), a plateau between 40 and 69 years, and a decline above age 70. Consequently, the relative risk of occurrence of a contralateral breast cancer will decline with age, because the background reference incidence rises with age.6-8 This pattern of risk would be compatible with the elimination of susceptible individuals with advancing age, or with a hormone-related influence on contralateral breast cancer, which may decline with menopause.9-11 The incidence of neoplasms other than breast, however, does not seem to be elevated in women diagnosed with breast cancer in this population.12 The Vaud dataset also indicates that the incidence of first breast cancer in women with subsequent contralateral neoplasm rises steeply below age 40, reaches a plateau between 40 and 69 years, and tends to decline thereafter, i.e. the slope of the age curve is substantially different from that of women without subsequent contralateral breast cancer, again pointing to the existence of a susceptibility in this subgroup of women.3, 4 The contributions of the Vaud Cancer Registry's staff are gratefully acknowledged. Fabio Levi Fax: +4121-3230303*, Lalao Randimbison*, Van-Cong Te*, Carlo La Vecchia , * Registre vaudois des tumeurs and Unité d'épidémiologie du cancer, Institut universitaire de ḿdecine sociale et préventive, Lausanne, Switzerland, Istituto di Ricerche Farmacologiche “Mario Negri”, Milan, Italy, Istituto di Statistica Medica e Biometria, Università Degli Studi di Milano, Milan, Italy

The relation between various types of fiber and oral, pharyngeal and esophageal cancer was investigated using data from a case-control study conducted between 1992 and 1997 in Italy. Cases were 271 hospital patients with incident, histologically confirmed oral cancer, 327 with pharyngeal cancer and 304 with esophageal cancer. Controls were 1,950 subjects admitted to the same network of hospitals as the cases for acute, nonneoplastic diseases. Cases and controls were interviewed during their hospital stay using a validated food frequency questionnaire. Odds ratios (OR) were computed after allowance for age, sex, and other potential confounding factors, including alcohol, tobacco consumption, and energy intake. The ORs for the highest vs. the lowest quintile of intake of oral, pharyngeal and esophageal cancer combined were 0.40 for total (Englyst) fiber, 0.37 for soluble fiber, 0.52 for cellulose, 0.48 for insoluble non cellulose polysaccharide, 0.33 for total insoluble fiber and 0.38 for lignin. The inverse relation were similar for vegetable fiber (OR = 0.51), fruit fiber (OR = 0.60) and grain fiber (OR = 0.56), and were somewhat stronger for oral and pharyngeal cancer than for esophageal cancer. The ORs were similar for the two sexes and strata of age, education, alcohol and tobacco consumption, and total non-alcohol energy intake. Our study indicates that fiber intake may have a protective role on oral, pharyngeal and esophageal cancer.

An innovative approach was used to define a low-risk diet for colorectal cancer from a multicentric case-control study of 1953 incident cases and 4154 hospital controls from Italy. A logistic regression model was fitted on the reported intake of five macronutrients, and the estimated coefficients were used to compute a diet-related logistic risk score (LRS). The mean of LRS within risk decile ranged from 0.89 to 1.86. Total energy intake and absolute consumption of each macronutrient increased with increasing LRS. In relative terms, however, starch intake showed an almost threefold increase across subsequent score levels, while a decline was observed for unsaturated fat, sugar and protein. Saturated fat consumption remained fairly stable in relative terms. When food groups were considered, bread and cereals dishes, cakes and desserts and refined sugar were positively associated, while the consumption of vegetables, fruit, fish, poultry and olive oils was inversely associated with LRS.

Weiderpass et al. [(1)][1] , on a record linkage cohort study, found an excess risk for in situ and invasive cervical cancer among alcoholic women. Little is known about the potential role of lower levels of alcohol consumption in the etiology of cervical cancer. In the interview study from the