Evidence to support a change in follow-up policy for patients with breast cancer: Time to first relapse and hazard rate analysis

The ideal follow-up for patients with cancer should be sensitive to the lik elihood of relapse, for prompt investigation and treatment if indicated, to gether with the support of patient confidence. The current British Associat ion of Surgical Oncologists guidelines for patients with breast cancer sugg est intensive follow-up, including 3-monthly clinic visits during the first 2 years. These recommendations place increasing demands on clinical resour ces. The combined outcome of screening, early detection, dedicated clinical services that emphasize rapid diagnosis and concomitant improved survival, have resulted in increasing absolute numbers of diagnosed breast cancer pa tients in follow-up clinics. This article examines the follow-up of breast cancer patients to determine if the convention should be adjusted to obtain more from current resources while maintaining equivalent patient care. The data on all patients with breast cancer attending one general oncology clinic were examined in order to determine the pattern of relapse. Analyses identified: (1) the time to relapse at any site and at specific sites; and (2) the prognostic significance of three factors for subsequent relapse, n amely nodal status, menopausal status and T stage at diagnosis. In 416 consecutive patients, the annual rate of relapse of breast cancer wa s found to increase progressively over the first 4 years. Nodal disease was the most important single variable as a predictor of relapse. The annual h azard rate for relapse for node positive patients in the first year was 5%; this increased to 10% and 14% in years three and four respectively. In con trast, in those patients who were node negative at diagnosis (302/416; 73%) , the hazard rate for relapse was 1% in year one, increasing to 5% in years three and four. Intensive early follow-up of breast cancer patients provides no clear clini cal gain for the great majority of patients, since early relapse is rare in the first year. The use of clinical funds and staff resources might be opt imized to focus clinical follow-up on those patients at risk of recurrence. We suggest that all patients should continue to be monitored and receive p sychological care through access to their general practitioner, skilled bre ast care nurses and specialized counsellors. Any patient at risk, or develo ping symptoms of relapse, would have immediate clinical access to the oncol ogist for diagnostic investigations. This strategy would optimize psycholog ical patient care and use the full backup of clinical resources during the prolonged period over which relapse becomes more probable.