Effect of Hormone Replacement Therapy in Breast Cancer Patients

Abstract & Commentary

Synopsis: Hormone replacement therapy (HRT) has been found to increase the risk of developing breast cancer and, for that and other reasons, it is often avoided in women previously diagnosed with breast cancer. Scientists from the University of Washington published the results of a sophisticated case-control study and concluded that HRT after breast cancer has no adverse effect on recurrence or mortality.

Source: O’Meara ES, et al. J Natl Cancer Inst. 2001;93:754-762.

The medical records of 400,000 group health Cooperative of Puget Sound HMO patients were searched for women with primary invasive breast cancer from 1977 to 1994, and 2755 were identified. Only those women without distant metastatic disease were selected. HMO pharmacy data were used to determine which subjects in the study group used HRT. Other factors, including parity, gravidity, age at first full-term pregnancy, age at menarche, age at and reason for cessation of menses, menopausal symptoms, hysterectomy, oophorectomy, smoking history, family history, height, weight, details of breast cancer treatment, and tumor characteristics were also extracted from the patients’ records. An HRT user was defined as any woman who filled 2 or more HRT prescriptions within any 6-month period after breast cancer diagnosis and prior to diagnosis of a recurrence.

There were 175 women who were classified as HRT users and were evaluable. Forty-three percent used vaginal preparations exclusively, 41% used only oral agents, and 16% used both. Adherence to the prescribed HRT regimen was assumed. Dose-equivalents of conjugated estrogen were calculated for esterified estrogen and ethinyl estradiol. Tubes of vaginal HRT were counted. Four non-HRT users from a pool of 698 women were then matched to each user based on age, year of diagnosis, and stage at diagnosis. Nonusers were still included even if they subsequently began HRT at some date later than that date chosen to match a user’s interval from the time of their breast cancer diagnosis to the entry point in the study. HRT was used before diagnosis by 68% of HRT users after diagnosis and by 48% of nonusers. Patients were followed for a median of 3.7 years for recurrence and 4.6 years for mortality.

Recurrences were diagnosed in 16 HRT users (9%) and 101 nonusers (15%). The unadjusted relative risk (RR) was 0.58, and the RR after adjusting for potential confounding features such as those mentioned above, was 0.50. Analysis by type of hormone replacement did not change the RR results. Five HRT users (3%) and 59 nonusers (8%) died during the follow-up period. The unadjusted RR was 0.31, and the adjusted RR was 0.34.

O’Meara and colleagues concluded from their results that women who used HRT after a diagnosis of breast cancer had lower risks of recurrence and death than nonusers. They stated that their results should be interpreted with caution given the limitations of the study.


Menopausal symptoms can result in significant discomfort and impaired quality of life for sufferers. Symptoms can be severe and prolonged. No agents offer efficacy equivalent to HRT with regard to amelioration of these symptoms. Given that estrogen has been associated with an increased risk of breast cancer, and tamoxifen, with its antiestrogen properties, has been shown to decrease breast cancer recurrences, there is great concern over whether HRT in breast cancer survivors is wise. For the most part, HRT is discouraged in the latter group. However, data are very limited. O’Meara et al used a detailed, meticulous approach in their case-control study and determined that HRT does not put breast cancer survivors at increased risk of new events. On the contrary, HRT appeared to offer a protective effect.

In an accompanying editorial, Cuzick suggests that, even if HRT is found to be neutral with respect to an effect on breast cancer prognosis, it would represent an important advance in patient management.1 Potentially, it would enable physicians to treat patients’ menopausal symptoms without fear of provoking breast cancer into recurring.

Col and associates from Brigham and Women’s Hospital in Boston performed a MEDLINE search for studies published from 1966 to 1999 and calculated RR values for 11 papers reporting on HRT following a diagnosis of breast cancer. Their objective was to combine all the existing data to establish the effect of HRT in this patient cohort. Their results coincided with the findings of O’Meara et al in that no significant increase in the risk of breast cancer recurrence was identified. O’Meara et al suggested that, although not conclusive, their results indicated that any risk associated with HRT must be of a limited magnitude.

Despite the data showing the absence of a deleterious effect of HRT on outcomes in breast cancer survivors, clinicians must continue to proceed with caution when counseling patients regarding the use of estrogens or estrogen derivatives. We are awaiting the results of 2 randomized trials that may offer further insights into this vexing issue.3,4


1 Cuzick J. J Natl Cancer Inst. 2001;93:733-734.

2 Col NF, et al. J Clin Oncol. 2001;19:2357-2363.

3 Marsden J, et al. Acta Obstet Gynecol Scand. 1997; 76(abstract suppl):22.

4 Vassilopoulou-Sellin R, et al. J Natl Cancer Inst. 1994; 6:153-159.