and United Kingdom; investment in further cohort studies seems wise, particularly with new formulations of the pill becoming available and changes in use patterns.

Ecological data describing breast cancer incidence and mortality and oral contraceptive distribution and use are sparse, and they are likely to be of limited utility in countries with high rates of breast cancer. Whether they can be fully exploited worldwide will depend on the quality of national information systems. Chapter 2 explores this opportunity more fully. The chapter also discusses the need to plan a fresh cohort study as soon as a new oral contraceptive preparation is brought into general use, as well as the opportunity to capture essential new information that arises in countries that have relatively low background rates of breast cancer and that decide to introduce the pill as a generally available contraceptive.

The potential contributions of existing animal and in vitro human tissue models to questions of oral contraceptives and breast cancer have not been fully exploited, perhaps because they are mostly being used by investigators who are working within their own disciplines to answer other specific questions. Opportunities must be provided for interdisciplinary research that concentrates the expertise of investigators who use these models on efforts to get answers about oral contraceptives and breast cancer. These opportunities are discussed in Chapter 3. The existing evidence from animal models (see Appendix D) has been influential in continuing the effort to explore the possible influences of oral contraceptives on breast carcinogenesis.

Beside the rising rates of breast cancer incidence there is a virtual absence of definitive information about the pathological processes of breast carcinogenesis. Available experimental data from studies of different mammals are conflicting. However, the sum of the data suggests that some of the steroids used in oral contraceptives in certain dosage patterns are capable of inducing tumors in experimental animals. Both increased tumor incidence and decreased latency have been observed. Given the widespread use of these compounds, even a small increase in risk for lifetime cancer incidence would be expected to have a substantial effect on human populations using these hormones. This suggests that a careful, formalized risk assessment should be conducted and compared with the human epidemiological data. The assumptions for this risk assessment should be explicitly defined and delineated.

A formal, four-step cancer risk assessment methodology (comprising hazard identification, hazard characterization, exposure assessment, and risk characterization) has been used extensively over the past decade to protect the public health from known or suspected



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement