Estradiol in Breast Cancer Treatment: Title and subTitle BreakReviving the Past

Despite a declining incidence, breast cancer remains the most common cancer among women in the United States.¹ More than two-thirds of all patients with breast cancer present with tumors that express estrogen receptors, progesterone receptors, or both, and the modulation of estrogen receptor signaling has been one of the most successful strategies for these patients. In recent years, the most commonly used forms of endocrine therapy have included the competitive inhibition of the estrogen receptors with an antiestrogen (selective estrogen receptor modulators [SERMs], eg, tamoxifen, or selective estrogen–receptor down-regulators, [SERDs], eg, fulvestrant) and the decrease in estrogen production from precursor steroid hormones using an aromatase inhibitor.

In this issue of JAMA, Ellis and colleagues² describe the findings of a preliminary phase 2 randomized study, evaluating the efficacy of 2 daily doses of oral estradiol (30 mg vs 6 mg). The study included 66 postmenopausal women with hormone receptor–positive metastatic breast cancer, who had received prior therapy with an aromatase inhibitor for at least 24 weeks for metastatic disease or for at least 2 years as adjuvant therapy. The authors reported a clinical benefit for 29% of patients (10/34) treated with the low-dose (6 mg) therapy and for 28% of those (9/32) treated with high-dose (30 mg) therapy. Grade 3 and 4 toxicities were significantly more common in patients receiving the higher dose of estradiol. The adverse effects associated with the 30-mg dose of estradiol led to treatment discontinuation in 4 of 32 patients and required a dose reduction in 4 patients, whereas only 1 patient withdrew from therapy at the lower dose for personal reasons.

Estradiol-related pharmacodynamic end points, including changes in the fluorodeoxyglucose–positron-emission tomography/computed tomography (FDG-PET/CT) uptake over 24 hours after estradiol initiation and a decrease in the expression of insulinlike growth factor (IGF-1), were observed in both groups. The increase in the specific uptake value by FDG-PET/CT in predefined tumor sites appeared predictive of response, although data were available for only 70% (46) of the patients. Based on the promising efficacy data and the acceptable toxicity profile observed with the low-dose estradiol, the authors suggest that further studies exploring this intervention should be pursued in larger randomized phase 3 trials.²

Synthetic estrogens were an integral part in the armamentarium of breast cancer for several decades. First synthesized in 1938, diethylstilbestrol (DES) was approved as a synthetic estrogen by the US Food and Drug Administration in 1941, and recommended as a treatment for breast cancer for postmenopausal women in 1960.³ Very high doses of estrogen were further tested in premenopausal women, albeit with more limited success.⁴ The use of estrogens for the treatment of breast cancer was largely empirical and perhaps counterintuitive because the benefits of estrogen withdrawal by oophorectomy, adrenalectomy, and hypophysectomy were well-recognized by that time as a treatment for breast cancer in both premenopausal and postmenopausal women by that time.^{5 - 7} Furthermore, formal testing of estrogen receptor expression was not yet available.

High doses of estrogens remained one of the few nonsurgical options for breast cancer treatment until the introduction of the antiestrogen tamoxifen in 1971.⁸ A small randomized phase 2 trial evaluating 2 doses of tamoxifen in postmenopausal women with advanced breast cancer and prior exposure to hormonal manipulations showed sustained activity with acceptable toxicity with this agent, which had been unsuccessfully developed as a contraceptive.⁹ Further studies comparing tamoxifen with DES showed that DES and tamoxifen elicited similar results in terms of response rates and time to tumor progression; however, tamoxifen treatment led to far fewer adverse effects and treatment discontinuations.¹⁰ The subsequent introduction of third-generation aromatase inhibitors and the approval of the SERD fulvestrant rendered synthetic estrogens a less desirable alternative, and high-dose estrogens have been rarely used in recent years. Furthermore, changes in practice patterns have shifted the use of chemotherapy more to the foreground, even for women with hormone receptor–positive tumors.

The interest in estrogens as therapy for metastatic breast cancer was rekindled when Lønning et al¹¹ reported a benefit of DES in patients with breast cancer who had progressed after receiving several lines of newer types of endocrine therapies. Furthermore, long-term follow-up of the original study comparing tamoxifen and DES suggested a survival benefit for women treated with DES compared with those treated with tamoxifen.¹²

The study presented by Ellis et al² is particularly intriguing because clinical benefits were observed with lower doses of estradiol. These findings suggest that in contrast to the toxicities observed with the 30-mg dose in this study and reported in previous studies, a lower dose of 6 mg of estrogen can be administered with acceptable toxicities and fewer profound effects on quality of life. The mean plasma levels in patients treated daily with 6 mg of estradiol were more comparable to estradiol levels observed during the normal menstrual cycle. The clinical benefits and the effects on the estrogen-associated biomarkers, including an estrogen-induced flare measured by FDG-PET/CT and a decrease in IGF-1, suggest that higher doses of estrogen may not be necessary to induce the desired effects in patients after prolonged aromatase inhibitor exposure.

A major challenge of hormone therapy is the definition of hormone therapy resistance in preclinical models as well as in patients. Hormone therapy resistance encompasses de novo resistance and emerging or acquired resistance, intact or suppression of estrogen receptor expression.^{13 - 17} The inclusion criteria in the study by Ellis et al² required estrogen receptor expression and a benefit from aromatase inhibitor treatment for at least 24 weeks. These criteria and the long interval from initial diagnosis (median, 82 months for the 30-mg group and 90 months for the 6-mg group) would suggest that these patients have hormone-sensitive disease with acquired rather than de novo resistance. Although it may be speculated that the prior exposure to an aromatase inhibitor sensitized cells to low-dose estrogens, this study was not designed to rule out a potential benefit from estrogen among women without prior aromatase inhibitor treatment earlier in their treatment course. However, this study² and the study by Lønning et al,¹¹ provide strong support for the concept of continued modulation of the estrogen receptor in patients with hormone-sensitive tumors. Preclinical and clinical data further suggest that sustained clinical benefit can be achieved by alternating between inhibitory and stimulatory estrogen receptor modulation, or even by withdrawal of therapy.^{13 ,15 ,18 - 21}

The randomized phase 2 design of the study by Ellis et al² has important strengths and limitations.²² In lieu of the more typical end points, such as progression-free or overall survival, randomized phase 2 trials use alternate end points including response rates or toxicity to determine whether a novel intervention shows sufficient promise for further testing. The smaller sample set limits the level of confidence, yet prevents unnecessary exposure to inactive interventions and ensures the comparability of the patient population.²³ This randomized phase 2 trial testing the response rate to 2 different doses of estrogen in postmenopausal women showed nearly identical clinical benefit rates between the low and the higher dose of estrogen, but major differences in toxicities between the 2 doses; however, given the wide confidence intervals, a difference in efficacy could have been missed. Nonetheless, the comparable efficacy in the setting of significant differences in toxicities between 2 doses supports further testing of low-dose estrogen in patients with breast cancer resistant to aromatase inhibitors.