Changes in survival after breast cancer: improvements in diagnosis or treatment?

We have compared 5-year survival rates in two cohorts of women diagnosed with breast cancer in Brisbane, Australia, between 1981–1984 and 1990–1994. Tumours diagnosed in the early 1990s were significantly smaller and less likely to have nodal involvement than those diagnosed 10 years earlier (P<0.0001). The size difference was particularly striking for women aged over 50 at diagnosis, those targeted for screening. Five-year survival was greater among women diagnosed in the 1990s (84% vs. 74%; hazard ratio (HR) 0.61, 95% confidence interval (CI) 0.46–0.81). After adjusting for the effects of tumour size and nodal status this difference was reduced, but women diagnosed more recently still showed improved survival (HR 0.75; 95% CI 0.56–1.01) and disease-free survival (HR 0.72; 0.56–0.92) at 5 years. This suggests that both earlier diagnosis and changes in breast cancer treatment have contributed to improved breast cancer survival.