Diet and ovarian cancer : A population-based cohort of 60 000 women

Abstract: Ovarian cancer represents the fourth leading cause of cancer-related mortality among women in Western countries. Hence, identification of risk factors amenable to modification, such as diet, is of considerable public health relevance as it could provide means for primary prevention of this malignancy. The purpose of this thesis was to prospectively examine the potential role of diet in the etiology of ovarian cancer, using data from the Swedish Mammography Cohort (SMC). The specific hypotheses were: (1) Intake of folate – a B vitamin involved in DNA synthesis and repair and in DNA methylation – is inversely associated with the risk of ovarian cancer, particularly among women who consume alcohol, which has anti-folate effects; (2) High exposure to lactose due to consumption of milk and milk products is related to an increased risk of ovarian cancer. The SMC is a population-based prospective cohort established between 1987 and 1990, when all women born between 1914 and 1948 and living in Uppsala and Västmanland Counties received a mailed questionnaire. Our analyses included 61 084 women in the SMC who were cancer-free and had at least one ovary at recruitment. Cox proportional hazards models were used to estimate rate ratios (RRs) with 95% confidence intervals (CIs), while adjusting for potential confounders. All statistical tests were two-sided. From March 1987 through June 2004, we ascertained 288 incident cases of invasive epithelial ovarian cancer, including 135 serous ovarian cancers. Dietary folate intake was inversely related to the risk of ovarian cancer. Compared with women in the lowest quintile (<150 μ g/day) of dietary folate intake, the multivariate RR for those in the highest quintile ( ³ 212 μ g/day) was 0.50 (95% CI, 0.31-0.82; P trend = 0.006). The inverse association between dietary folate intake and risk of ovarian cancer was stronger among women who consumed more than 20 g (approximately the amount in 1-2 drinks) of alcohol per week (RR, 0.24; 95% CI, 0.09-0.64) than among women who consumed 20 g or less of alcohol per week (RR, 0.72; 95% CI, 0.40-1.30). We observed a positive association of lactose intake with risk of serous ovarian cancer but not with other subtypes. The multivariate RR of serous ovarian was 2.0 (95% CI, 1.2-3.4; P trend = 0.01) for the highest versus the lowest quartile of lactose intake. High consumption of milk, the main source of lactose, was also associated with a greater risk of serous ovarian cancer. The multivariate RR for women who drank two or more glasses of milk per day was 1.9 (95% CI, 1.1-3.1; P trend = 0.03) compared with women who never or seldom drank milk. In conclusion, our findings from a large population-based prospective cohort provide evidence for an inverse relationship between dietary folate intake and risk of ovarian cancer, especially among women who drink alcohol. Furthermore, we found support for the hypothesis that a high lactose intake increases the risk of ovarian cancer, at least of the serous subtype. Considering the possible public health implications of these results, the associations of folate and lactose intakes with risk of ovarian cancer warrant further research.