An meta-analysis of 21 epidemiological studies found some support for the hypothesis that high intake of milk products and lactose is associated with increased ovarian cancer risk.
Since a positive correlation between ovarian cancer risk and the consumption of milk products and lactose was first reported in 1989, many epidemiological studies have further examined the relationship. However, these studies have yielded conflicting conclusions.
To better understand the uncertain relationship, researchers led by Susanna C. Larsson of the National Institute of Environmental Medicine at the Karolinska Institute in Sweden, conducted a meta-analysis of relevant case-control and cohort studies.
The researchers sought reports that offered data from a case-control, or cohort study on the association between intakes of milk, yogurt, cheese or lactose, and incidence of or mortality from epithelial ovarian cancer. Studies also had to present results as an odds ratio, or relative risk, with 95 percent confidence intervals. The researchers accepted three prospective cohort studies, and 18 case-control studies and performed a meta-analysis to determine associations between consumption and cancer risk.
Their analysis found notable differences between case-control and cohort studies. Case-control studies showed low-fat milk consumption negatively associated, and whole milk consumption positively associated, with the risk of ovarian cancer, but offered no support for the involvement of lactose in the development of ovarian cancer. By contrast, prospective cohort studies indicated that high intakes of milk may increase the risk of ovarian cancer. They also revealed a 13 percent increase in ovarian cancer risk with a daily increase of 10 grams of lactose, the approximate amount in one glass of milk. When case-control and cohort studies were considered in combination, yogurt consumption was associated with a non-significant increase in cancer risk, while cheese was not associated with risk.
The differences between the findings of case-control studies and those of cohort studies could be explained by a number of factors: selection bias, recall bias or changes in dietary practices after cancer diagnosis. They might also be due to the time interval between diet assessment and illness, since cohort studies may record dietary practices decades before illness occurs, while case-control studies assess diet around the time of diagnosis. Other limitations of this study include the observational nature of the assessed studies, imprecise measurements of diet, and publication bias.
Of note, the two studies that examined histological subtypes of ovarian cancer found that the associations with milk and lactose intakes were confined to serous ovarian cancer, leading researchers to advise, "future studies should consider specific subtypes of ovarian cancer, and the interrelationship between intakes of dairy foods and lactose, genetic polymorphisms, and ovarian cancer risk."
"In conclusion," they write, "prospective cohort studies, but not case-control studies, support the hypothesis that high intakes of dairy foods and lactose may increase the risk of ovarian cancer."
Source: International Journal of Cancer, 2005