Changing pattern of age-specific breast cancer incidence in the Swiss canton of Geneva.

Hormone replacement therapy (HRT) use declined sharply after mid-2002, when the Women's Health Initiative trial reported an association between breast cancer occurrence and HRT. Hypothesized mechanism behind this association is that HRT promotes growth of pre-existing small tumors, leading to earlier tumor detection. We evaluated the impact of the sudden decline in HRT use on age distribution of breast cancer in Geneva. We included all incident breast cancer cases recorded from 1975 to 2006 at the Geneva cancer registry. We calculated mean annual incidence rates per 100,000 for 2 year periods for three age groups and assessed temporal changes by joinpoint regression. We compared age-specific incidence curves for different periods, reflecting different prevalence rates of HRT use. After increasing constantly between 1986 and 2002 among women aged 50-69 years [annual percent change (APC): +4.4, P < 0.0001], rates declined sharply after 2003 (APC: -6.0; P = 0.0264). Age-specific breast cancer rates changed dramatically with changes in prevalence of HRT use. During low HRT prevalence, breast cancer incidence increased progressively with age, when HRT prevalence was reaching its maximum (1995-2002), higher rates were seen in 60- to 64-year-old women, with a concomitant decrease in risk among elderly. After the sudden decline in HRT use, the incidence peak diminished significantly and incidence increased again with age. Following the abrupt decline in HRT use in Geneva, breast cancer incidence rates among post-menopausal women decreased considerably with striking changes in age-specific incidence rates before, during and after the peak in HRT prevalence.

Remarkable change in age-specific breast cancer incidence in the Swiss canton of Geneva and its possible relation with the use of hormone replacement therapy.

BACKGROUND: This article aims to explain the reasons for the remarkable change in age of breast cancer occurrence in the Swiss canton of Geneva. METHODS: We used population-based data from the Geneva cancer registry, which collects information on method of detection, stage and tumour characteristics since 1975. For patients diagnosed between 1997-2003, we obtained additional information on use of hormone replacement therapy from a large prospective study on breast cancer. Using generalized log linear regression analysis, we compared age-specific incidence rates with respect to period, stage, oestrogen receptor status, method of detection and use of hormone replacement therapy. RESULTS: In the periods 1975-1979 and 1985-1989, breast cancer risk increased with age, showing the highest incidence rates among women aged >or= 85 years. From 1997, the age-specific incidence curve changed completely (p < 0.0001), showing an incidence peak at 60-64 years and a reduced incidence among elderly women. This incidence peak concerned mainly early stage and oestrogen positive cancers and was exclusively observed among women who ever used hormone replacement therapy, regardless whether the tumour was screen-detected or not. CONCLUSION: The increasing prevalence of hormone replacement therapy use during the 1990s could explain the important change in age-specific breast cancer incidence, not only by increasing breast cancer risk, but also by revealing breast cancer at an earlier age.