The natural history of breast cancer: implications for a screening strategy.

In the data base of the Institut Gustave-Roussy, the clinical courses of over 7000 patients treated since 1951 are registered. For 3000 of them treated prior to the introduction of adjuvant chemotherapy, the follow-up ranges from 15 to 32 years. For all patients, the size of the primary tumor, its histologic grade, and the number of involved axillary nodes at the time of initial treatment were registered prospectively. This series of patients was used to analyze the relationship between the size of the primary tumor and the probability of distant metastatic spread, taking into account other prognostic variables. A simulation model of the natural history of breast cancer was built which satisfactorily fits the available data on breast cancer screening. The aim of this paper is to use this model to assess the influence of: a) the time interval between mammographies and b) the diagnostic ability of the screening program, on the proportion of patients with distant metastases. The results show that the proportion of patients with distant metastases at the time of diagnosis increases gradually from approximately 25% for an interval of 1 year to 40% for an interval of 5 years. Moreover, the proportion of patients with metastases is critically influenced by the size of the tumors which can be detected; for example, for a 2-year interval the proportion of patients with metastases increases from 32% for tumors screened of 1 cm in diameter to 40% for tumors of 1.7 cm in diameter.