Abstract
Small, occult, undiagnosed breast cancers are found at autopsy in up to 15.6 % of women dying from unrelated causes with an average of 7 % from eight separate studies. The mammographic detection threshold of breast tumors ranges from 0.88 to 1.66 cm in diameter based on the patient's age. Tumor growth rates, expressed as "effective doubling times," vary from 10 to >700 days. We previously reported two models, based on iterative analysis of these parameters, to describe the biologic behavior of undiagnosed, occult breast tumors. Our models facilitate interpretation of the Women's Health Initiative (WHI) and antiestrogen breast cancer prevention studies. A nude mouse xenograft model was used to validate our assumption that breast tumors grow in a log-linear fashion. We then used our previously reported occult tumor growth (OTG) and computer-simulated tumor growth models to analyze various clinical trial data. Parameters used in the OTG model included a 200-day effective doubling time, 7 % prevalence of occult tumors, and 1.16 cm detection threshold. These models had been validated by comparing predicted with observed incidence of breast cancer in eight different populations of women. Our model suggests that menopausal hormone therapy with estrogens plus a progestogen (E + P) in the WHI trial primarily promoted the growth of pre-existing, occult lesions and minimally initiated de novo tumors. We provide a potential explanation for the lack of an increase in breast cancer incidence in the subgroup of women in the WHI who had not received E + P prior to randomization. This result may have reflected a leftward skew in the distribution of occult tumor doublings and insufficient time for stimulated tumors to reach the detection threshold. Our model predicted that estrogen alone reduced the incidence of breast cancer as a result of apoptosis. Understanding of the biology of occult tumors suggests that breast cancer "prevention" with antiestrogens or aromatase inhibitors represents early treatment rather than a reduction in de novo tumor formation. Our models suggest that occult, undiagnosed tumors are prevalent, grow slowly, and are the biologic targets of a hormone therapy in menopausal women and of antiestrogen therapy for prevention.