Reply to The Women's Health Initiative; hormone replacement therapy; and Surveillance, Epidemiology, and End Results data.

Abstract

Our commentary1 presented the hypothesis that changes in menopausal hormone practices beginning in 2002, as initially described by Ravdin et al,2 may be a major contributor to differential breast cancer national time trends between Black and White women. Wachtel proposed that his analysis provided “evidence against the theory that HRT [hormone replacement therapy] increases the risk of breast cancer.” Our analysis of the influence of menopausal hormone therapy (HT) on breast cancer is not based on breast cancer time trends but rather on the Women's Health Initiative randomized, placebo-controlled trials evaluating conjugated equine estrogens plus medroxyprogesterone acetate (CEE+MPA) (16,608 patients), which demonstrated a statistically significantly increased incidence of breast cancer among women randomized to active treatment and on a companion randomized trial of CEE alone (10,739 patient), which demonstrated an unexpectedly statistically significant decrease in the incidence of breast cancer.3 Preclinical findings have provided the biological rationale for these results.4-6 After initial publication of the results of the CEE+MPA trial in 2002, approximately two-thirds of US users of combined estrogen and progestin (combined HT) formulations, as well as approximately one-third of US users of estrogen-alone preparations, ceased HT use.7 Since that time, several groups have reported substantial and sustained reductions in population breast cancer risks in the United States.8, 9 In our view, statistical modeling of US time trends, including joinpoint modeling, did not contribute significantly to evidence that HT alters breast cancer risk.1 Instead, investigation of the association between use of HT and breast cancer incidence at the regional8 or participant level9 provided more compelling evidence regarding HT cessation and breast cancer incidence. DeSantis and Siegel have suggested that our hypotheses are based on arguments that “are inconsistent with several recent publications concerning breast cancer incidence.” They correctly state that our use of long-term breast cancer trend data are “based on data from the 9 oldest Surveillance, Epidemiology, and End Results (SEER) registries.” It is unclear to us whether DeSantis and Siegel are arguing that analyses using a different set of cancer registries would yield a conflicting hypothesis concerning racial group time trends in breast cancer incidence. Part of the argument made by DeSantis and Siegel is based on their assertion that the increased risk of breast cancer associated with HT is limited to patients with hormone receptor–positive subtypes, citing 3 observational studies. However, in the Women's Health Initiative randomized trial, an increased incidence of breast cancers among patients treated with CEE + MPA was not limited to those women with hormone receptor–positive cancers.3 In contrast to another assertion made by DeSantis and Siegel, there is information regarding longer-term breast cancer risk among women in a combined HT trial who infrequently used systemic hormones after the intervention. In that study, a median of 5.6 years of CEE + MPA use sustained a statistically significant increase in breast cancer incidence through a median cumulative follow-up of 13 years initially3 and now 20 years.10 On a population level, as combined HT use continues to occur at a low level, many women now entering menopause no longer experience the long-term increase in the risk of breast cancer observed with combined HT use. As we outlined in our commentary,1 among Black women, the prevalence of combined HT use was only a small percentage of the use noted in White women.11 Therefore, Black women would not have experienced the sustained reduction in breast cancer incidence noted among newly menopausal White women who chose not to use combined HT. This differential pattern of HT use between Black and White women supports our hypothesis with regard to how a long-observed breast cancer incidence advantage for Black women largely has dissipated over the past 20 years. Supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health (contracts N01WH22110, 24152, 32100-2, 32105-6, 32108-9, 32111-13, 32115, 32118-32119, 32122, 42107-26, 42129-32, and 44221). Rowan T. Chlebowski has received honorarium and consulting fees from Novartis, AstraZeneca, and Genentech; consulting fees from Puma and Immunomedics; and personal fees from Pfizer for work performed as part of the current study. Garnet L. Anderson and Ross L. Prentice have received grants from the National Heart, Lung, and Blood Institute of the National Institutes of Health for work performed as part of the current study. Aaron K. Aragaki made no disclosures.