Estrogen Replacement Therapy for Breast Cancer Patients

The discussions and debates about the use of estrogen replacement therapy (ERT) in women with breast cancer often seem to ignore or at least leave unnoted the extensive data supporting the general premise that increased, but physiologic levels of estrogens are associated with poorer survival in postmenopausal women with breast cancer. Dr. Colditz summarizes various lines of evidence bolstering this general premise, providing us with some needed lessons about the complexities of interpreting epidemiologic studies and about human cancer biology. Particularly illuminating are his discussion of the biases in ERT-breast cancer causation studies and his exploration of the reasons for the apparently better survival in ERT users who develop breast cancer.

More Evidence on the Importance of Physiologic Estrogen Levels

I would add two sets of evidence about the importance of physiologic levels of estrogens in premenopausal women in order to emphasize (1) the absence of differences between premenopausal and postmenopausal breast cancers, and (2) the importance and usefulness of descriptive epidemiologic data to clinical decision-making. First, while a woman’s long-term risk of breast cancer is reduced following a full-term pregnancy before age 30 to 35, her short-term risk is increased.[1] I suggest that this epidemiologic-causation observation is relevant to the debate about the impact of pregnancy on recurrence in women with breast cancer, in the same way that ERT-breast cancer causation data are relevant to ERT treatment decisions in women with breast cancer.

Second, two large studies have clearly demonstrated better survival in women who are diagnosed with breast cancer at ages 45 to 49 or 35 to 49, respectively, than in those who are either younger or older at diagnosis.[2,3] It is difficult to escape the conclusion that the menopause undergone by most of these women in the early years after diagnosis contributes to this better outcome. And, again, as a general premise, ovarian ablation is effective in premenopausal women with breast cancer.

Heterogeneity of Breast Tumors

A second general premise that is important in the debate about ERT in postmenopausal women with breast cancer is that breast tumors are heterogeneous with respect to tumor markers sensitive to hormonal change. Most tumors studied by immunohistochemical means for estrogen- or progesterone-receptor proteins have some cells that stain positive.[4] Thus, it appears that the hormone sensitivity of a tumor is often a quantitative rather than a qualitative issue. Although the patterns of recurrence over time differ in patients with different levels of hormone-receptor proteins, the overall frequencies of recurrence are similar. Therefore, the take-home message is as Dr. Colditz has argued: Recognizing the overall contribution of estrogen to disease progression, we should be chary about the use of ERT in women with breast cancer.

Applying the Message to Individual Patients

Of course, the application of this message to individual patients should be based on a thoughtful review of their particular risks and benefits. The main focus of concern should be on women with bothersome vasomotor and central nervous system disturbances. (As Dr. Colditz suggests, various clinical strategies can be used when cardiovascular or osseous risks are an issue.) For these bothersome symptoms, I try (in sequence of usefulness): clonidine, vitamin E, diazepam, and phenobarbital preparations—all of which are modestly helpful at best—and then consider megestrol alone in women whose symptoms are refractory to these agents. When this approach proves to be inadequate, I use ERT in the lowest dose, with a plan to taper the dose and stop therapy over 1 to 2 years

Contrary to the widely promoted program of concomitant use of tamoxifen (Nolvadex) and megestrol,[5] I never prescribe these two hormonal preparations together. This stance is based on laboratory data demonstrating a synergistic effect of this combination on the growth of breast cancer cells.[6] In particular, I considered that hormone replacement therapy recipients who develop breast cancer actually undergo a hormonal therapy via withdrawal of their ERT at diagnosis; in other words, that stopping ERT is, in itself, a significant treatment.

References:

1. Lambe M. Hsieh C, Tricopoulos D, et al: Transient increase in risk of breast cancer after giving birth. N Engl J Med 331:5-9, 1994.

2. Adami HO, Malker B, Holmberg L, et al: The relation between survival and age at diagnosis in breast cancer. N Engl J Med 315:559-563, 1986.

3. Host H, Lund E: Age as a prognostic factor in breast cancer. Cancer 57:2217-2221, 1986.

4. Remmele W, Schicketanz KH: Immunohistochemical determination of estrogen and progesterone receptor content in human breast cancer: Computer-assisted image analysis (QIC score) vs subjective grading (IRS). Pathol Res Pract 189:862-866, 1993.

5. Loprinzi Cl, Michalak JC, Quella SK, et al: Megestrol acetate for the prevention of hot flashes. N Engl J Med 331:347-352, 1994.

6. Iino Y, Wolf DM, Langan-Fahey SM, et al: Reversible control of oestradiol-stimulated growth of MCF-7 tumors by tamoxifen in the athymic mouse. Br J Cancer 64:1019-1024, 1991.