Increasing Age Predicts Breast Cancer Mortality in Certain Women

A woman’s risk of dying as a direct result of breast cancer increases with age, according to results of a recent study of postmenopausal women with hormone receptor positive disease [JAMA. 2012;307(6):590-597]. This risk increased regardless of other factors. TEAM (Tamoxifen, Exemestane, Adjuvant, Multinational) was a randomized, phase 3, multinational, open-label study of 9766 postmenopausal women with hormone receptor positive breast cancer (estrogen, progesterone, or both) that took place between January 2001 and January 2006. Without an upper age limit, the TEAM researchers were able to examine the association between age and disease-specific mortality. Participants were eligible if they had completed local therapy and did not have metastatic disease. They were categorized into 3 age groups: <65 years, between 65 and 74 years, and ≥75 years. The primary end point was disease-specific mortality, defined as time from randomization to death due to breast cancer. Secondary end points included other-cause mortality and breast cancer relapse. Participants were randomized to receive either exemestane (25 mg/day) for 5 years, or tamoxifen (20 mg/day) for 2.5 to 3 years followed by exemestane (25 mg/day) for another 2 to 2.5 years. The study population included 5349 patients <65 years of age at diagnosis (55%, n=5349), 3060 patients between the ages of 65 and 74 years (31%, n=3060), and 1357 patients aged ≥75 years (14%, n=1357). Results showed the cumulative incidence of death due to breast cancer increased from 5.7% for patients <65 years of age, 6.3% for patients aged 65 to 74 years, to 8.3% for patients ≥75 years of age. The cumulative incidence of death from other causes was 1.6%, 4.9%, and 14.6%, respectively. As a proportion of all-cause mortality, increasing age was associated with fewer deaths due to breast cancer (77.5% for those <65 years of age, 56.3% for those between 65 and 74 years, and 36.3% for participants ≥75 years of age; P<.001). However, results of the univariate Cox regression analysis showed increasing age to be related to a higher risk of disease-specific mortality, with a hazard ratio (HR) of 1.12 (95% confidence interval [CI], 0.94-1.34) for patients 65 to 74 years of age and 1.66 (95% CI, 1.34-2.06; P<.001) for those ≥75 years of age. After adjusting for tumor and treatment characteristics, results showed increasing age again to be associated with disease-specific mortality. The HR for patients between 65 and 74 years of age was 1.25 (95% CI, 1.01-1.54) and for patients ≥75 years of age, 1.63 (95% CI, 1.23-2.16; P<.001). An analysis to test the robustness of the age cut points also showed an increased risk of disease-specific mortality for each 10-year increase in age (univariate HR per 10 years, 1.20; 95% CI, 1.10-1.31; P<.001; multivariable HR per 10 years, 1.21; 95% CI, 1.08-1.36; P=.001). There was a significant age-associated increase in larger tumors (P<.001); therefore, another analysis to adjust for tumor size (in centimeters as opposed to T category) was performed, which yielded similar results (HR for participants between 65 and 74 years of age, 1.25; 95% CI, 1.01-1.55, and for those ≥75 years of age, 1.62; 95% CI, 1.22-2.14; P=.003). Because other-cause mortality increases with age, additional survival analyses were done to take into account competing mortality as well as comorbidity. Results of these showed HRs to be similar to those mentioned previously. After categorizing age into 7 groups, researchers found disease-specific mortality to be similar for patients <70 years of age, but after 70 years of age, mortality rose stepwise with increasing age. Breast cancer relapse also occurred more frequently with increasing age (HR, 1.07, 95% CI, 0.91-1.25 for those between 65 and 74 years of age and 1.29, 95% CI, 1.05-1.60 for those ≥75 years of age; P=.06). As the study included only postmenopausal women with estrogen or progesterone receptor positive tumors (or both), results may not necessarily apply to all breast cancer patients. Additionally, the researchers acknowledged competing mortality is likely to be higher in the general population.