A 13-year follow-up of patients with breast cancer presenting to a District General Hospital breast unit in southeast England.

The purpose of this clinical cohort study was to examine long-term survival in groups of screen-detected and symptomatic breast cancer patients attending a specialist breast unit, and to determine the factors affecting this survival. A total of 3239 breast cancer cases (1252 screen-detected and 1987 symptomatic) diagnosed between 1989 and 2002 were followed up until the end of 2002, and breast cancer-specific survival was examined in the screened and symptomatic groups, in relation to patient age, tumour size, nodal status and histological grade. Long-term survival in this clinical cohort was high. Relative survival values in the symptomatic group were similar to population-based estimates for southeast England derived from data held at the Thames Cancer Registry, whilst survival values in the screening group were 11-12 percentage points higher at 4 years after diagnosis and onwards. Ten-year relative survival estimates were 88% in the screening group and 77% in symptomatic cases. In groups of comparable age, the difference in breast cancer-specific survival at 10 years was 19 percentage points. Survival was significantly related to tumour characteristics (size, nodal status and grade), but not to morphological tumour type or deprivation score of the subject. After adjustment for other factors, the difference in survival between the two groups was much attenuated and became statistically non-significant. The higher survival seen in the screening group can be almost entirely accounted for in terms of the detection of early stage and lower grade tumours which screening provides.