Hormone Replacement Therapy After a Diagnosis of Breast Cancer in Relation to Recurrence and Mortality

Abstract

Although hormone replacement therapy (HRT) is generally not recommended for women who have been treated for breast cancer, some patients have used HRT because of its beneficial impact on bone density and lipid profiles. This study was undertaken to investigate the impact of the use of HRT on recurrence and survival of women who have had breast cancer. The records of 2755 women aged 35 to 74 years who had a diagnosis of invasive breast cancer between 1977 and 1994 were identified in the cancer surveillance system of a large health maintenance organization. The 174 patients who had used HRT after being treated for breast cancer serve as the study population. They were each matched to four control subjects who corresponded in age, stage of disease, and year of diagnosis. All study subjects were disease-free when HRT was initiated, and control subjects were disease-free for a corresponding interval of time since diagnosis. Oral, vaginal, or combination HRT use was fairly equally distributed among study subjects (38, 41, and 16%, respectively). The average duration of use was 15 months. Nearly 80% of the women used unopposed estrogens. In the course of the study, 16 of the 174 women (9%) who used HRT had breast cancer recurrence, compared with 101 of the women (15%) who did not use HRT. The rate of breast cancer recurrence was 9% (16 of 174) among the women who used HRT and 15% (101 of 696) among nonusers. Calculated in numbers of person-years, the rate of recurrence in users was 17 per 1000, compared with 30 per 1000 in nonusers. The relative risk of developing a recurrence in women who had ever used HRT was 0.58 (95% CI = 0.34-0.98). This risk was similar in other analyses adjusting for possible contributing variables, including bilateral oophorectomy, hysterectomy, mastectomy, and use of tamoxifen. No appreciable change was seen in opposed versus unopposed estrogen users or method of HRT delivery (oral vs. vaginal). The relative risk approached 1 with greater duration of use and larger accumulated HRT dose. Five HRT users (3%) and 59 nonusers (8%) died of breast cancer during the study period, yielding a breast cancer mortality rate, expressed in person-years, of 5 per 1000 for HRT users and 15 per 1000 for nonusers. The relative risk of dying of breast cancer for HRT users was 0.31 (95% CI = 0.13-0.78) and remained essentially the same when adjusted for obesity, positive lymph nodes, and oral or vaginal HRT use. Similarly, the total death rate from any cause, which was 10% among users and 17% among nonusers (relative risk = 0.54, 95% CI = 0.33-0.90), did not change appreciably with adjustments. Women with estrogen receptor-negative tumors who used HRT had an unadjusted relative risk of breast cancer recurrence of 0.31 (95% CI = 0.10-0.98), compared with 0.81 (95% CI = 0.30-2.17) for women with estrogen receptor-negative tumors who did not use HRT.