Trends in breast cancer incidence and mortality in the United States: implications for prevention.

While debate continues regarding short-term changes in breast cancer incidence and the impact of screening on mortality, a long-term view of trends in incidence and mortality may better inform our understanding of the changing patterns of disease and ultimately guide in population-based prevention. Although many factors have influenced breast cancer incidence over the past seven decades, some have played more prominent roles at various times. Changing reproductive patterns, greater longevity, and post-menopausal hormone (estrogen + progesterone) were important in the steady increase before 1980, while mammographic screening, probably in conjunction with escalating combined estrogen + progesterone use, played dominant roles in the post-1980 surge. Accruing evidence also indicates that the rapid drop in 2003 was mostly due to a sharp decline in estrogen + progesterone use. The most paradoxical observation relates to the divergence in incidence and mortality trends most noticeable when mortality rates started to decline shortly after the surge in incidence rates started in 1980. In addition to the dynamic changes in risk factor profiles, the divergence reflects wider uptake of screening mammography, better characterization of tumor biology, and improvements in treatment. The rise in incidence rates over the past three decades is due to an increase in estrogen receptor positive (ER+) tumors, which respond favorably to treatment. On the other hand, the incidence of estrogen receptor negative (ER-) tumors, which respond poorly to hormonal therapy, has been decreasing for almost three decades. Furthermore, widespread adoption of screening mammography has led to tumors being diagnosed at earlier stages when treatment is effective and advances in treatment have ensured adoption of targeted and better tolerated therapies. To achieve long-term success in the primary prevention of breast cancer, a greater understanding of factors responsible for the decrease in ER- tumors is essential. In addition, improving the sensitivity of breast cancer screening to facilitate earlier detection of tumors with very aggressive phenotypes would go a long way in bridging the divergence between incidence and mortality.