The appropriate follow-up of primary breast cancer patients is a controversial topic: an intensive surveillance is the rule for many oncologists, but the beneficial effect of this attitude has never been demonstrated. Two consensus conferences focused on this issue, both discouraged the routine use of laboratory tests, chest X-ray, radionuclide bone scan or liver echography which involve a high financial cost. For these routine investigations the sensitivity to detect recurrence is less than 5%. History and physical examination remain the best methods to detect a recurrence, and mammogram is helpful for detecting recurrent disease in the treated breast or a new primary cancer in the controlateral breast, which both are curable. The purpose of intensive follow-up is earlier detection of recurrence in order to treat as early as possible, but two randomized studies failed to demonstrate a survival benefit in 2600 patients. However with current treatments, only a follow-up study including a very large number of patients could detect a survival difference. The demonstration of an intensive follow-up benefit is similar in screening studies where it is recommended to use the most predictive, the less toxic and the less expensive test. Serum C.A. 15.3 level which best anticipates diagnosis in follow-up is one of the best candidates to be used in a randomized study.