Abstract
Endocrine therapy is among the most powerful tools currently available for prevention and treatment of breast cancer, whether in the adjuvant or metastatic setting. Since the “modern” era of endocrine therapy, when tamoxifen was first approved in the adjuvant setting, a bewildering variety and variance in endocrine therapy have emerged. This chapter endeavors to discuss (1) the history and development of endocrine therapy, (2) general strategies for targeting this pathway, and (3) key therapeutic agents. We then turn to a discussion of endocrine therapy in (4) the management of ductal carcinoma in situ, (5) the management of early stage hormone receptor–positive breast cancer, and finally (6) in the advanced and metastatic setting. Key considerations for inclusion in this review were studies that have advanced our knowledge of optimal clinical use of endocrine therapy and novel treatments to overcome resistance to these agents.
Keywords
hormone receptors, endocrine therapy, aromatase inhibitors, tamoxifen, ovarian function, suppression, fulvestrant, resistance.
History of Endocrine Therapy in Breast Cancer
Linking the Course of Advanced Breast Cancer to Female Reproductive Organs
In 1836, Sir Astley Cooper made one of the earliest known observations suggesting a role for endocrine therapy in treatment of breast cancer. He observed that advanced breast cancer appeared to wax and wane during the course of a woman’s menstrual cycle. In 1889, Schinzinger proposed making younger women “older” by removing their ovaries, noting that younger women with breast cancer had more aggressive disease than older women. In 1895, Beatson extended this rationale into the arena of treatment by performing a bilateral oophorectomy for a woman who had developed recurrent breast cancer involving the chest wall 6 months after mastectomy. He reported that her chest wall disease resolved within 8 months of the bilateral oophorectomy and the patient remained disease free for 4 years. Boyd subsequently reported a series of 54 patients who underwent bilateral oophorectomy as treatment for advanced breast cancer. Approximately one-third of patients had tumor regression and improved overall survival. This ushered in an era of oophorectomy for the treatment of advanced breast cancer, although oophorectomy was eventually replaced by ovarian ablation techniques such as irradiation.
Recognizing Hormone Dependency for Certain Human Tumors
Huggins and Bergenstal reported the beneficial effects of orchiectomy in men with prostate cancer, and subsequently, Huggins and Hodges showed that the effect of orchiectomy was mediated by reducing testosterone levels. Block and associates contributed by demonstrating that the basal level of estrogen production was not reached for several months after ovarian radiation, correlating with the fact that breast cancers often did not regress until several months after ovarian radiation. When synthetic corticosteroids became available in the early 1950s, bilateral surgical adrenalectomy also became feasible as a means of removing other sources of steroid hormones. In an early report by Huggins and Bergenstal, three of six patients with advanced breast cancer appeared to benefit from bilateral adrenalectomy. Other investigators subsequently demonstrated objective response rates of 30% to 40% in advanced breast cancer patients after bilateral adrenalectomy.
Understanding Estrogen Action and Developing Antiestrogens
Modern drug development originated with the pioneering studies of Dodds, Lawson, and Noble in the 1930s and their discovery of the nonsteroidal estrogen diethylstilbestrol (DES). In 1944, Haddow observed that women with advanced breast cancer responded to high-dose estrogen. The women who responded tended to be older, but it was unclear why only some women responded to endocrine therapy. Similarly, responses to bilateral adrenalectomy had been noted to occur in premenopausal women, who had previously responded to oophorectomy. Manipulation of female hormone levels was clearly unsuccessful in certain populations, but insight into the mechanism was lacking.
This clinical observation was correlated with mechanism when Jensen and Jacobson observed that radiolabeled estradiol localized to estrogen target tissues such as the uterus, vagina, and pituitary gland. They proposed that a receptor must be present in these tissues to regulate response to estradiol. The estrogen receptor (ER) was subsequently identified. ER assays were then developed to predict which breast cancer patients would respond to endocrine therapy. The response rate to endocrine therapy was 30% to 40% in unselected patients, compared with 60% or more in women with a positive ER assay.
In 1958 Lerner, Holthaus, and Thompson reported the biological properties of the first nonsteroidal antiestrogen, MER-25. The compound was found to be an antiestrogen in all species tested. MER-25 was initially studied as a contraceptive in laboratory animals. Unfortunately, the large doses needed for MER-25 to work were associated with unacceptable central nervous system side effects. A successor compound, MRL-41, or clomiphene, was a more potent antiestrogen and an effective antifertility drug in animals, although it paradoxically induced ovulation in subfertile women. Clomiphene demonstrated modest activity in the treatment of advanced breast cancer, but further development stopped after the introduction of tamoxifen.
The Modern Era of Endocrine Therapy
Similar to clomiphene, Harper and Walpole demonstrated that tamoxifen, a selective estrogen receptor modulator (SERM), had antiestrogenic and antifertility properties. Tamoxifen was evaluated in a number of clinical scenarios, including as a contraceptive. The first successful use of tamoxifen in treating advanced breast cancer was reported in 1971, with 10 of 46 patients (22%) demonstrating responses to therapy. The response rates were similar to those with DES; however, side effects were significantly less with tamoxifen. Thus tamoxifen became the endocrine therapy of choice for advanced breast cancer in the 1970s. In 1986, tamoxifen was approved for adjuvant treatment of postmenopausal women with node-positive breast cancer. In 1990 tamoxifen was approved as an adjuvant treatment for pre- and postmenopausal women with node-negative disease.
General Strategies for Targeting the Hormonal Axis
Hormone Assays
Approximately two-thirds of breast cancers express female hormone receptors (HR): either ER or progesterone receptor (PR), or both ER and PR. The decision to choose one endocrine therapy over another must take into consideration the comparative efficacy, ease of administration, and toxicity of therapy, as well as the menopausal status of the patient. There are multiple techniques available for determining the ER and PR status of tumors. The ER can be measured by a ligand-binding method after isolation from tumor sample. This method has multiple shortcomings, including expense, requirement of fresh frozen tissue, and use of radioactive reagents. The development of monoclonal antibodies to specific to ER and PR and immunohistochemistry (IHC) techniques provided an alternative means of determining ER/PR status. IHC is currently the most commonly used technique for assessing ER and PR status and overcomes many of the problems associated with ligand binding assays. One of the challenges of IHC is that the results are subjective and do not adequately quantitate the level of ER or PR expression. Newer techniques, such as reverse transcriptase polymerase chain reaction (PCR), tissue microarrays, and nanotechnology, are being investigated as alternatives to IHC, with improved quantitation of ER and PR levels.
Predictive Power of ER/PR Status
The HR status of a tumor determines the likelihood that a patient will respond to endocrine therapy: 75% to 80% of patients with tumors positive for both ER and PR will respond to an initial endocrine therapy. The response rates to endocrine therapy are lower for ER-positive/PR-negative tumors and ER-negative/PR-positive tumors at 25% to 30% and 40% to 45%, respectively. However, patients with very low levels of ER and/or PR expression may still benefit from endocrine therapy. Such findings led to the current American Society of Clinical Oncology (ASCO) guidelines that even patients with very low levels of ER or PR expression (≥1%) be recommended to receive endocrine therapy. Patients whose tumors do not express either ER or PR at all typically do not benefit from endocrine therapy, although there is always a concern about false negativity of the assays especially when evaluating ER and PR status from a bone biopsy.
The likelihood of expressing ER and/or PR increases with age and postmenopausal status. However, several other clinical factors other than menopausal status can be used to predict response to endocrine therapy. Patients with prolonged disease-free intervals after their initial diagnosis are more likely to remain HR positive. Additionally, patients with metastatic disease in soft tissue or bone, but not in visceral or central nervous system sites, are more likely to respond to endocrine therapy.
Key Therapeutic Agents
As Table 70.1 demonstrates, there are now multiple drugs to prevent the development of breast cancer as well as to prevent and treat breast cancer recurrence. For a few decades, tamoxifen was considered the standard of care for first-line endocrine therapy for all women with metastatic breast cancer and was the only therapy in the adjuvant setting. This changed with the development of aromatase inhibitors (AIs). AIs prevent the peripheral conversion of androstenedione into estrogen, resulting in decreased levels of circulating estrogen. AIs are not effective in the management of premenopausal breast cancer patients (in whom the ovaries are the main source of estrogen) although they are used in premenopausal women for fertility purposes. Early nonselective AIs, such as aminoglutethimide, were poorly tolerated by patients and replaced by newer AIs developed in the 1990s. The selective AIs (anastrozole, letrozole, and exemestane) have proved to be active drugs for postmenopausal women with hormone-sensitive breast cancer.
| Drug | Class | Menstrual Status Studied | Breast Cancer–Related FDA Indication(s) | Key Side Effects |
|---|---|---|---|---|
| Tamoxifen | SERM |
|
|
|
| Raloxifene | Post | Breast cancer prevention |
| |
| Toremifene | Post | First-line treatment of HR+ MBC |
| |
| Anastrozole | AI | Post |
|
|
| Exemestane |
| |||
| Letrozole |
| |||
| Fulvestrant | SERD | Post | Treatment of advanced HR+ breast cancer after progression on antiestrogen therapy | Abnormal liver enzymes |
| Estradiol | Estrogen | Metastatic breast cancer, palliative treatment |
| |
| Fluoxymesterone | Androgen | Metastatic breast cancer, palliative treatment: |
| |
| Megestrol acetate | Progesterone | Metastatic breast cancer, palliative treatment |
|
Selective Estrogen Receptor Modulators
Tamoxifen
Efficacy.
Tamoxifen is currently US Food and Drug Administration (FDA)-approved for the treatment of all stages of hormone-responsive (aka HR-positive) breast cancer and for the prevention of breast cancer in high-risk women. As noted earlier, the efficacy of tamoxifen proved to be equivalent to that of androgens or high-dose estrogens such as DES in postmenopausal women, but the side effects of tamoxifen were mild in comparison. The efficacy of tamoxifen in both premenopausal and postmenopausal women has been demonstrated in multiple clinical trials. Clinical benefit (defined as complete response plus partial response plus stable disease) is observed in 50% to 60% of HR-positive cancers. Duration of response ranges from 12 to 18 months, although select women may benefit for longer. For premenopausal women, tamoxifen is the first-line treatment of choice for hormone-sensitive advanced disease (see discussion later in the chapter on ovarian function suppression).
Side Effects.
Tamoxifen has antiestrogenic effects on some tissues including the breast and has partial estrogenic effects elsewhere in the body. This complex mechanism of action results in side effects of treatment both beneficial and detrimental. In postmenopausal women treated with tamoxifen, clinical studies have shown an increase in trabecular bone density and a trend toward decreased loss of cortical bone density. The National Surgical Adjuvant Breast and Bowel Project (NSABP) P-1 chemoprevention trial demonstrated fewer osteoporotic fracture events in women who received 5 years of tamoxifen compared with placebo; however, the results did not reach statistical significance. This reduction is mainly limited to postmenopausal women.
Tamoxifen has been shown to have beneficial effects on the lipid profile. In adjuvant breast cancer trials, tamoxifen significantly lowers total cholesterol, mainly due to its effect on low-density lipoprotein (LDL) cholesterol. Tamoxifen also lowers fibrinogen, lipoprotein(a), and homocysteine, all factors that contribute to cardiovascular risk. However, until recently, no trial had demonstrated a reduction in cardiac events in patients taking tamoxifen. Extended follow-up of the Swedish tamoxifen adjuvant trial demonstrated reduced mortality from coronary heart disease in patients receiving 5 years of adjuvant tamoxifen, compared with those receiving 2 years of treatment.
Tamoxifen has been associated with an increased incidence of endometrial carcinoma. The relative risk of endometrial cancer in the tamoxifen-treated women from the NSABP P-1 prevention trial was 2.5. The increased risk was predominantly seen in women over age 50 in whom the relative risk was 4. All the endometrial cancers seen in the tamoxifen-treated women were International Federation of Gynecology and Obstetrics (FIGO) stage I. The tumors were of good prognosis, and none of the women treated with tamoxifen died from endometrial cancer. There was also an increased incidence of deep venous thrombosis in the tamoxifen-treated women in the NSABP P-1 trial. The relative risk of pulmonary embolism in the tamoxifen group was 3.0.
Metabolism.
The cytochrome P450 enzyme CYP2D6 catalyzes the formation of endoxifen and low to absent CYP2D6 activity due to common genetic variation significantly lowers the plasma concentration of endoxifen. Goetz and coworkers first reported an association between endoxifen levels and benefit from tamoxifen by demonstrating that the presence of the CYP2D6*4 variant allele was an independent predictor of breast cancer relapse in postmenopausal women. They also showed that women with the variant allele had a lower incidence of hot flashes while taking tamoxifen. Two additional studies of CYP2D6 and tamoxifen response had shown contradictory results, although differences in study populations made comparisons difficult. The routine use of CYP2D6 testing to predict tamoxifen benefit is not currently recommended.
Raloxifene
Efficacy.
Raloxifene or keoxifene is another SERM that was initially developed as a treatment for breast cancer. Raloxifene has a shorter half-life than tamoxifen and has less estrogenic effects on the endometrium in preclinical models. Raloxifene was evaluated in patients with tamoxifen-refractory metastatic breast cancer. There were no partial or complete responders, with only one patient achieving a minor response, indicating significant cross-resistance between raloxifene and tamoxifen. Another study evaluated a high-dose schedule of raloxifene at 150 mg twice daily as first-line therapy for HR-positive metastatic breast cancer. The response rate was less than 20%, lower than what would be expected in this first-line setting. On the basis of these results, further development of raloxifene as a treatment for breast cancer was discontinued.
The Study of Tamoxifen and Raloxifene (STAR) trial was designed as a follow-up prevention trial to the NSABP P-1 trial. Postmenopausal women (n = 19747) who were considered high risk for developing breast cancer based on the Gail model were randomized to receive 5 years of tamoxifen or raloxifene. The rate of invasive breast cancers was not significantly different between the two treatment groups and was lower than what would have been expected in this population without treatment. Interestingly, there were fewer noninvasive cancers in the raloxifene-treated patients compared with the patients who received tamoxifen. The reason for this difference in noninvasive cancers is unclear but suggests the intriguing possibility that raloxifene is not preventing breast cancers but rather treating occult breast cancer. This theory was initially put forth to explain the results noted in the NSABP P-1 trial with tamoxifen. However, the sustained decrease in breast cancers noted with longer follow-up, and the fact that tamoxifen was also noted to decrease benign breast lesions places this theory into question. On the basis of the results of the STAR trial, raloxifene was approved as a chemopreventive in postmenopausal women at high risk of developing breast cancer. However, the fact that raloxifene has been developed in postmenopausal women only limits its use as a chemopreventive, given the rise of AIs as chemopreventives.
Side Effects.
In the STAR trial, the rate of endometrial cancer was lower in patients treated with raloxifene compared with tamoxifen. The fact that this difference did not quite reach statistical significance may be partly explained by the fact that approximately 50% of women on the STAR study had hysterectomies before study entry. The Multiple Outcomes of Raloxifene Evaluation study randomized almost 8000 women with osteoporosis to placebo or one of two doses of raloxifene and showed no increase in the rate of endometrial cancer in patients treated with raloxifene. Raloxifene was associated with a significantly reduced incidence of thromboembolic effects compared with tamoxifen, although there was no difference in the rate of cerebrovascular accidents. The rate of osteoporotic fractures was similar between the two groups.
In summary, raloxifene is currently approved for the prevention and treatment of osteoporosis and for the prevention of breast cancer in postmenopausal women. Raloxifene should not be used as an alternative to tamoxifen in patients with invasive breast cancer or ductal carcinoma in situ (DCIS) because there is currently no data at present to support its use in these settings. Additionally, raloxifene has not been evaluated in premenopausal women and should not be used as a chemopreventive in these women.
Toremifene
Efficacy.
Toremifene is a chlorinated derivative of tamoxifen that is currently approved in the United States as an alternative to tamoxifen in the first-line treatment of hormone-responsive metastatic breast cancer. Toremifene was noted to have minimal activity in tamoxifen-refractory metastatic breast cancer, indicating almost complete cross-resistance between the two SERMs. Five trials have compared toremifene at various doses with tamoxifen in the first-line treatment of hormone-receptor positive metastatic breast cancer. A meta-analysis of these trials demonstrated equal efficacy and toxicity between the two SERMs. Toremifene is a reasonable alternative to tamoxifen in this setting. Of note, toremifene, unlike tamoxifen, has not been evaluated in premenopausal patients. Therefore, although toremifene can be considered a reasonable alternative to tamoxifen in postmenopausal patients with metastatic breast cancer, the widespread use of AIs in this group of patients limits its clinical importance.
Toremifene has been compared with tamoxifen as adjuvant therapy in patients with early-stage breast cancer. Postmenopausal patients (n = 1480) with node-positive early-stage breast cancer were randomized between tamoxifen 20 mg and toremifene 40 mg daily for 3 years. At a follow-up of just over 3 years, the rate of recurrence was similar between patients treated with the two agents, 23% and 26% for patients treated with toremifene and tamoxifen, respectively ( p = .31).
Side Effects.
There was no significant difference in the side effects associated with either treatment, including the incidence of endometrial cancer.
Other Selective Estrogen Receptor Modulators
Other SERMs are in clinical use, such as ospemifene (Osphena), which is approved for women experiencing moderate to severe dyspareunia due to menopause. However, none of the SERMs in clinical practice or development appear to offer a significant advantage over tamoxifen in the treatment of breast cancer, with the possible exception of endoxifen, which is still being studied.
Aromatase Inhibitors
The primary source of serum estrogens in postmenopausal women is circulating androgens, mainly in the adrenals but with a small contribution from the postmenopausal ovaries. These androgens (mainly androstenedione) are converted into estrone, followed by reduction to estradiol (aromatization of circulating testosterone into estradiol contributes to a lesser degree). By preventing the conversion of androgens to estrogens via inhibition of the enzyme responsible, AIs lower serum estrogen levels. Over time, a series of AIs have been developed and studied. Fadrozole, formestane, aminoglutethimide, and vorozole were all studied and used to treat advanced breast cancer but are no longer clinically available. As a result, this section focuses on anastrozole, letrozole, and exemestane, which are currently available and in use. Initial studies conducted to evaluate the efficacy of third-generation AIs in patients with metastatic disease in the second-line setting compared these agents to megestrol acetate.
Steroidal Versus Nonsteroidal Aromatase Inhibitors
Anastrozole and letrozole are oral, substrate analogs of androstenedione, the normal substrate of the aromatase enzyme, and reversibly inhibits the enzyme. Exemestane is an oral, irreversible steroidal inhibitor of the aromatase enzyme. All three selective third-generation AIs potently inhibit aromatase activity and thus significantly reduce serum estrogen levels, although letrozole appears to be the most effective at reducing estrogen levels. However, the clinically significant threshold for estrogen reduction is unknown. A consequence of aromatase inhibition can be an increase in serum androgen levels.
Early Aromatase Inhibitor Trials
Buzdar and associates reported the results of two parallel, multicenter trials involving 764 women with metastatic, hormone-responsive breast cancer who had progressed after treatment with tamoxifen. Two doses of anastrozole, 1 mg and 10 mg daily, were evaluated compared with megestrol acetate (40 mg four times daily). At a median follow-up of 31.2 months, overall survival was significantly improved with anastrozole (1 mg: 26.7 months; 10 mg: 25.5 months vs. 22.5 months). This difference was statistically significant for the 1-mg dose of anastrozole. The improvement in survival was particularly interesting because there was no difference in response rates or time to progression between the AI and megestrol at either dose of anastrozole. Dyspnea, hypertension, weight gain, and vaginal bleeding were among the major side effects that were at least two times more common with megestrol acetate. Subsequent trials with exemestane and letrozole versus megestrol demonstrated relatively similar results: improved efficacy and less toxicity.
A pooled analysis of the trials using second- and third-generation AIs in comparison with megestrol acetate after tamoxifen failure was conducted to determine whether there was a difference in efficacy between the two types of endocrine therapy. In this analysis, there was no significant difference in overall outcomes, and the efficacy was determined to be equivalent. However, there were notably more side effects experienced by patients treated with megestrol acetate, particularly increased weight gain, dyspnea, and peripheral edema.
Aromatase Inhibitors Side Effects
Overall these agents are well tolerated. In general, there is little difference in the side effect profiles between AIs, although individual-level variations clearly exist in side effects experienced on one AI versus another. In clinical trials, the major side effects reported were hot flashes, musculoskeletal pain, vaginal dryness, and headache. AIs do not carry the association with thromboembolic disease seen with tamoxifen, as studies involving patients without a cancer diagnosis comparing AIs with placebo showed no increase in events. Additional potential long-term side effects include bone loss and effects on lipid profiles.
The ATAC (Arimidex, Tamoxifen Alone or in Combination) trial analysis also found the incidence of musculoskeletal disorders was significantly higher in the anastrozole-treated patients (35.6% vs. 29.4%, p < .0001). Conversely, the risk of endometrial cancer ( p = .007), venous thromboembolic disease ( p = .0004), vaginal bleeding or discharge, and hot flashes ( p < .001 for all three) was lower in patients treated with anastrozole than tamoxifen. A bone substudy was conducted within the ATAC population. This substudy found the mean decrease in bone mineral density (BMD) for women treated with anastrozole was 6.08% in the lumbar spine and 7.24% in the total hip. Women treated with tamoxifen conversely, had a median gain in BMD of 2.77% in the lumbar spine and 0.74% in the total hip. Notably, among the patients with normal BMD at baseline, there were no cases of osteoporosis.
The rate of fractures seen in the BIG 1-98 trial was also significantly increased in women treated with letrozole compared with tamoxifen-treated patients (5.7% vs. 4%; p < .001). In the MA.17 companion study to evaluate the effect of letrozole on bone mineral density, 226 patients (122 letrozole and 104 placebo) were enrolled and prospectively evaluated for baseline BMD and changes in BMD over time. At 24 months, patients had significantly increased BMD loss in the total hip (3.6% vs. 0.71%; p = .044) and in the lumbar spine (5.35% vs. 0.70%; p = .008) compared with those receiving placebo.
Adverse effects on lipid profiles have been suggested in some studies, but a clinically significant increase in risk of cardiovascular disease has not been seen in the major studies of AIs, although a meta-analysis suggests that there may be a small but statistically significant increase in cardiovascular events. A study evaluated the effect of short-term therapy with 16 weeks of letrozole on plasma lipid profiles. This study did find a statistically significant increase in total cholesterol and LDL as well as unfavorable changes in the total cholesterol/HDL and LDL/HDL ratios. Another study of patients treated with exemestane or placebo found a significant decrease of 6% to 9% ( p < .001) in HDL levels and a 5% to 6% ( p = .004) decrease apolipoprotein A1 levels. A slight increase in serum homocysteine levels was also noted. All of these changes were reversed within a year of withdrawing AI therapy. A substudy of the MA.17 trial evaluated differences in plasma lipid profiles between 347 women enrolled in the study, from the letrozole (n = 183) and placebo (n = 164) arms, respectively. The study found no durable, significant change in plasma lipid profiles of patients treated with up to 36 months of letrozole compared with those receiving placebo.
Selective Estrogen Receptor Downregulators
Fulvestrant is a steroidal analog of 17-beta-estradiol. It binds to the ER, preventing receptor dimerization and thus effectively downregulating the ER, as shown in both preclinical and clinical studies. Fulvestrant is administered as an intramuscular injection because of poor oral bioavailability. Currently the only available agent in this class, fulvestrant was originally administered as a monthly intramuscular injection at a dose of 250 mg. However, later studies suggested that 500 mg monthly was more effective, with a 4.1-month improvement in overall survival. Unfortunately, this result also causes some difficulties interpreting the results of earlier studies using 250 mg monthly. A double-blind randomized trial compared the efficacy of fulvestrant 250 mg to anastrozole in 400 women with advanced breast cancer whose disease has progressed on prior endocrine therapy. Fulvestrant was shown to be equivalent to anastrozole for time to progression (hazard ratio 0.92; p = .92), overall response rate (17.5% in both arms), and clinical benefit 42.2% vs. 36.1%, p = .26). Duration of response in patients who had a clinical response was significantly longer in the fulvestrant arm (19 months vs. 10.8 months). A second study with a similar design enrolled 451 patients. In this study, fulvestrant 250 mg was as effective as anastrozole as second-line endocrine therapy for metastatic breast cancer, with regard to time to progression, overall response, clinical benefit rate, and duration of response. It was on the basis of these trials that fulvestrant has been approved by the FDA for use in women with metastatic breast cancer that has progressed on previous endocrine therapy.
Fulvestrant has also been evaluated as a potential therapy for patients who have progressed on an AI. A phase II study conducted by the Swiss Group for Clinical Cancer Research (SAKK 21/00) evaluated two separate groups of patients, those who were AI-responsive (n = 70) and AI-resistant (n = 20). In this study, all patients had previously received an AI, 84% had also been treated with tamoxifen or toremifene. Patients were treated with 250 mg of fulvestrant monthly until progression. Clinical benefit was obtained in 28% of patients with AI-responsive disease and 37% of patients with AI-resistant disease. Time to progression was similar in both groups at 3.6 and 3.4 months, respectively.
Both fulvestrant and exemestane have been evaluated as options in patients who have progressed on treatment with a nonsteroidal AI. A study compared fulvestrant to exemestane in this setting. Patients with metastatic hormone-responsive breast cancer who had experienced recurrence or progression on treatment with a nonsteroidal AI were recruited and 693 eligible patients were enrolled. Patients were randomized to treatment with fulvestrant 250 mg injected monthly or exemestane 25 mg daily. Two-thirds of patients had been previously treated with two or more endocrine therapies. Median time to progression was 3.7 months for both groups. For fulvestrant and exemestane, the response rates (7.4% vs. 6.7%, p = .736) and clinical benefit rates (32.2% vs. 31.5%, p = .853) were similar. The median duration of benefit for fulvestrant was 9.3 months and for exemestane was 8.3 months. On the basis of this study, fulvestrant 250 mg is equivalent to exemestane as a treatment option for patients with metastatic disease who have progressed after therapy with a nonsteroidal AI. However, as noted here (see also “Advanced and Metastatic Breast Cancer: Treatment”), 500 mg monthly appears superior to 250 mg monthly.
Fulvestrant is a good treatment option for patients with metastatic hormone-sensitive breast cancer that has progressed on previous tamoxifen therapy. Additionally, fulvestrant is active in patients who progressed on treatment with an AI. More about fulvestrant for the treatment of metastatic disease can be found in “Advanced and Metastatic Breast Cancer: Treatment.”
Fulvestrant Side Effects
In general, fulvestrant is well tolerated. Aside from injection site discomfort, hot flashes, elevations of liver enzymes, and arthralgias are possible.
Summary
Other agents such as high-dose estradiol, androgens, and progesterone agents are currently seldom used in the treatment of metastatic breast cancer, although they are included in Table 70.1 for the sake of completeness.
The most recent changes in endocrine therapy have been the inclusion of additional drugs to overcome primary and/or secondary resistance (reviewed in “Advanced and Metastatic Breast Cancer: Treatment”), and further insights into the optimal duration of therapy in the adjuvant setting (reviewed in “HR-Positive Invasive Breast Cancer: Adjuvant Treatment”) and the benefits of ovarian function suppression for premenopausal women (reviewed in “HR-Positive Invasive Breast Cancer: Adjuvant Treatment”).
Ductal Carcinoma in Situ
DCIS is noninvasive breast cancer, which remains confined by the basement membrane of the mammary duct. According to the American Joint Committee on Cancer 7th edition staging criteria, it is classified as stage 0 breast cancer. Currently, DCIS accounts for 20% to 30% of mammographically identified breast cancers compared with less than 3% of all newly diagnosed breast cancers before mammograms. Local treatment of DCIS involves either breast conservation typically followed by radiation or mastectomy. After definitive local treatment of DCIS, there continues to be a risk for in-breast recurrences as well as contralateral new primary breast cancer. Because the majority of DCIS are HR-positive, endocrine therapy has been evaluated to reduce this risk. NSABP B-24 is a double-blind randomized controlled study of tamoxifen in women with DCIS who had completed lumpectomy and whole breast radiation therapy. In this trial, 1804 pre- or postmenopausal women with resected tumors were randomized to tamoxifen 10 mg (n = 902) or placebo (n = 902) twice daily for 5 years. Stratification factors included age (≤49 or >49) presence of LCIS, and method of detection (mammogram, clinical examination, or both). Margins could be microscopically involved with tumor. The primary end point was the occurrence of invasive or noninvasive tumors in the ipsilateral or contralateral breast. At 5 years, the tamoxifen group had fewer breast cancer events than the placebo group (8.2 vs. 13.4%, hazard ratio 0.63; 95% confidence intervals [CI] 0.47–0.83; p = .0009). Subsequently, ER and PR were evaluated in 732 of the tumors: 76% had ER-positive DCIS. Patients with ER-positive DCIS on tamoxifen had experienced a significant reduction in any breast cancer event (31% vs. 20%; hazard ratio 0.58; 95% CI 0.42–0.81; p = .0015) after a median follow-up of 14.5 years. After multivariate analysis, tamoxifen reduced the risk to time to any breast cancer as first event with a hazard ratio of 0.64 (95% CI 0.48–0.86; p = .003) in the ER-positive group. No difference was seen in the ER- group. Although this was an unplanned retrospective, subset analysis of B-24, the isolation of benefit to ER-positive DCIS is supported by data from large adjuvant endocrine trials for invasive breast cancer.
The UK/ANZ randomized phase III trial used a 2 × 2 factorial design, in which women (n = 1701) with locally excised DCIS were randomized to radiation, tamoxifen, no adjuvant treatment or both. Tamoxifen was administered 20 mg daily for 5 years. The primary end point of the study was any new breast cancer event, including invasive ipsilateral new events, DCIS and contralateral disease. After a median follow-up of 12.7 years, tamoxifen reduced the risk of incidence of all new breast events with a hazard ratio 0.71 (95% CI 0.58–0.88; p = .002), but had no effect on ipsilateral invasive disease ( p = .8).
More recently, data have emerged regarding AIs in the setting of DCIS. The randomized, placebo-controlled phase III NRG Oncology/NSABP B-35 trial compared tamoxifen with anastrozole in postmenopausal women with HR-positive DCIS treated with lumpectomy and radiation. Women (n = 3104) in this trial were randomized and stratified by age (<60 vs. ≥60) to tamoxifen 20 mg (n = 1552) once daily or anastrozole 1 mg (n = 1552) once daily for 5 years. The primary end point was breast cancer–free interval (BCFI), defined as time from randomization to any breast cancer event including local, regional, or distant recurrence or a diagnosis of contralateral invasive breast cancer or DCIS. With a median follow-up of 9 years, the 10-year point estimates for BCFI were 89.2% for tamoxifen and 93.5% for anastrozole (hazard ratio 0.73; p = .03). A significant interaction was seen between treatment and age group ( p = .04) with benefit of anastrozole seen only in those women aged less than 60 years. Ten-year overall survival was no different in the two groups at 92.1% and 92.5% ( p = .48).
HR-Positive Invasive Breast Cancer: Adjuvant Treatment
Although the use of endocrine therapy has significantly improved survival for HR-positive breast cancer, many questions remain for this subset of breast cancer. Traditionally, the recommendation for adjuvant chemotherapy for ER and/or PR-positive tumors was based on patient age, tumor stage, and grade. However, this approach led to overtreatment of a large portion of women. Furthermore, it is now clear that HR-positive tumors can be divided into two groups, luminal A and luminal B, which have differential prognoses and response to therapies. On the basis of this information, additional tests are now available to aid decisions about adjuvant chemotherapy. Furthermore, resistance to endocrine therapy continues to be a key challenge. Intrinsic or de novo resistance to endocrine therapy exists in nearly 50% of HR-positive tumors. Tumors can also acquire resistance during endocrine therapy. Here we describe the currently available tests that can be used to aid decisions about adjuvant chemotherapy and duration of endocrine therapy.
Predictive Tools for Use of Adjuvant Chemotherapy in HR-Positive Breast Cancer
21-Gene Recurrence Score (Onco type DX Assay)
The 21-gene recurrence score (determined using the Onco type DX assay; Genomic Health, Redwood City, CA) measures the expression of 16 cancer-related genes and 5 reference genes in paraffin-embedded tissues using a quantitative PCR approach. The genes are primarily related to proliferation, invasion, and HER2 or estrogen signaling. This assay reports scores as low risk (<18), intermediate risk (18–30), or high risk (≥31), although the TAILORx and RxPONDER trials used different cutoffs for recommending use of chemotherapy. Reviewed here are the studies supporting the 21-gene recurrence score as a prognostic test for endocrine-treated ER-positive breast cancers and as a predictive test for adjuvant chemotherapy.
The prognostic significance of the 21-gene recurrence score assay was analyzed in a retrospective prospective substudy of patients enrolled in NSABP B-14. In B-14, patients with node-negative, ER-positive tumors were randomized to receive tamoxifen versus placebo without chemotherapy. Among the tamoxifen-treated patients, cancers with a high-risk recurrence score had a significantly worse rate of distant recurrence and overall survival. Inferior breast cancer survival with a high recurrence score was also confirmed in a case-control series of node-negative, tamoxifen-treated patients. The 21-gene recurrence score has also been demonstrated to be prognostic in tamoxifen-treated postmenopausal women with node-positive, ER-positive cancers. In this retrospective subset analysis of the SWOG 8814 trial, patients with low recurrence score cancers had a significantly improved disease-free and overall survival even when stratified for number of nodes involved. Furthermore, the 21-gene recurrence score has similar prognostic results for the aromatase inhibitors. The ATAC trial compared adjuvant tamoxifen to anastrozole in postmenopausal women with HR-positive breast cancers. The retrospective subset analysis of the 21-gene recurrence score on tumors from patients enrolled in this study demonstrated lower rates of distant recurrence with lower recurrence scores for both node-negative and node-positive patients. The prognostic value of the 21-gene recurrence score was similar in tamoxifen- and anastrozole-treated patients.
Beyond the prognostic value of the 21-gene recurrence score in tamoxifen-treated patients, the predictive utility of this assay has also been evaluated. In NSABP B-14, a comparison of the placebo and tamoxifen-treated patients demonstrated that the 21-gene recurrence score was prognostic for 10-year distant recurrence-free survival in both groups. It also predicted benefit from tamoxifen in cancers with a low- or intermediate-risk recurrence scores. In contrast, there was no benefit from the use of tamoxifen, compared with placebo, in cancers with high-risk recurrence scores. These data are intriguing and suggest that high-risk recurrence scores may predict for resistance to tamoxifen, and perhaps to other endocrine agents. Interestingly, the genes involved in the 21-gene recurrence score assay include ER, PR, HER2, and other genes associated with proliferation and invasion.
However, the greatest utility for the 21-gene recurrence score may be its ability to aid in decision-making regarding adjuvant chemotherapy in patients with node-negative and node-positive, ER-positive breast cancers. In the NSABP B-20 trial, adjuvant tamoxifen alone was compared with tamoxifen with chemotherapy in patients with node-negative tumors. A retrospective study of the 21-gene recurrence score in a subset of these tumors demonstrated that the benefit of chemotherapy was restricted to patients with breast cancers with higher recurrence scores. Similarly, in node-positive tumors, chemotherapy benefit was only seen in those with high 21-gene recurrence scores in the SWOG 8814 trial that compared adjuvant tamoxifen to tamoxifen plus chemotherapy.
Prospective studies are now underway to answer these questions more definitively. The TAILORx trial includes women with node-negative, HR-positive and HER2-negative tumors measuring 0.6 to 5 cm. The 21-gene recurrence score cutoffs were changed to low (0–10), intermediate (11–25), and high (≥26). Patients with high-score tumors were assigned to chemotherapy and endocrine therapy. Those with intermediate scores were randomized to chemotherapy or no chemotherapy followed by endocrine therapy. The results of these cohorts are still pending. However, results from the 1626 prospectively followed patients with low-risk tumors assigned to endocrine therapy alone demonstrated excellent outcomes with a 5-year risk of invasive disease-free survival of 93.8% and overall survival of 98%. RxPONDER is evaluating women with one- to three-node positive, HR-positive, HER2-negative tumors. In the RxPONDER trial, women with 21-gene recurrence scores of 0 to 25 were randomized to adjuvant chemotherapy and endocrine therapy versus endocrine therapy alone. Those with scores of 26 or higher were assigned to chemotherapy and endocrine therapy. This study recently completed accrual. Of note, neither of these studies is investigating the lack of benefit of endocrine therapy in cancers with high recurrence scores. Finally, the role of the 21-gene recurrence score in understanding the risk for late recurrence and the optimal duration of endocrine therapy is being explored.
PAM-50
The PAM-50 (Prosigna) is a quantitative PCR assay initially used to identify the intrinsic breast cancer subtypes (luminal A, luminal B, HER2-enriched, and basal-like). A newer Risk of Recurrence (ROR) score has been developed that includes this 50-gene expression profile using special weighting of a set of proliferation-associated genes as well as tumor size. In a series of node-negative or node-positive ER-positive breast cancers treated with adjuvant tamoxifen, the ROR score demonstrated prognostic utility. In addition, the ROR was evaluated for its prognostic significance for distant recurrence in tumor samples from the previously described ATAC trial. ROR correlated better with time to distant recurrence compared with the 21-gene recurrence score. However, it provided relatively similar information to IHC4 (ER/PR/HER2 and ki67). The prognostic utility of the ROR for distant recurrence was also demonstrated in an analysis of the ABCSG-8 study. In this study, postmenopausal women with ER-positive breast cancer received either adjuvant tamoxifen or anastrozole without chemotherapy. Those patients with a low ROR tumor score were found to have a very low risk for distant recurrence with 10-year DRFS of 96.7% (95% CI 94.6%–98%). More recently, the ROR was evaluated in a combined analysis of the ATAC and ABCSG-8 studies to explore the ability of this test to identify individuals at risk for late recurrence of breast cancer. Of 2137 women who did not have recurrence 5 years after diagnosis, the ROR score was prognostic beyond a clinical treatment score (including nodal status, tumor size, grade, age, and treatment for the risk of distant recurrence in years 5 to 10. A limitation of the PAM-50 data is the current lack of validation as a predictive biomarker. It has yet to be analyzed in randomized studies of chemotherapy with endocrine therapy versus endocrine therapy alone, or in trials comparing extended endocrine therapy to no extended therapy.
70-Gene Assay (MammaPrint)
The MammaPrint (Agendia, Amsterdam, The Netherlands) is a 70-gene DNA microarray assay that was developed for early-stage breast cancer. Initially, it was only available for fresh tissue analysis, but recent advances in RNA processing now allow for this analysis on formalin-fixed, paraffin-embedded (FFPE) tissue. MammaPrint scores are classified as low or high risk and are associated with 5-year risk for distant recurrence prognosis for node-negative and node-positive breast cancers. Comparisons of outcomes by MammaPrint for nonrandomized adjuvant chemotherapy with endocrine therapy or endocrine therapy alone cohorts of HR-positive breast cancers have reported that only high scores benefit from chemotherapy. The utility of MammaPrint for adjuvant chemotherapy decision-making is being explored in the MINDACT phase III trial in which patients with early-stage breast cancer (node-negative or up to three nodes positive) are treated based on clinicopathologic analysis and genomic risk using the MammaPrint 70 gene signature. Patients who had cancers that are high risk on the basis of both genomic and clinicopathologic analysis receive chemotherapy, in addition to endocrine and HER2-directed therapy where appropriate; patients who had cancers that are low risk by both criteria do not receive chemotherapy; patients with discordant results on either clinicopathologic or genomic criteria were randomized to receive chemotherapy or not. Almost 7000 patients were enrolled with 1800 having high-risk cancers and 2745 having low-risk cancers, using clinicopathologic and genomic criteria. In the discordant groups, almost 600 were deemed low risk by clinicopathologic criteria but high risk by genomic analysis, and 1550 were deemed high risk by clinicopathologic criteria but low risk by genomic analysis. The use of genomic analysis rather than clinicopathologic criteria spared 14% of patients chemotherapy. As was noted in the TAILORx trial, patients with cancers that were low risk by both criteria had extremely favorable 5-year outcomes without the use of chemotherapy, with distant metastasis–free and overall survivals of 98% and 98%, respectively. Interestingly, there was no benefit for the use of chemotherapy in the two discordant groups, including for cancers that were deemed genomically high risk.
ki67
ki67 is a nuclear nonhistone protein, the expression of which varies in intensity throughout the cell cycle. ki67 has been used as a measurement of tumor cell proliferation. A large meta-analysis demonstrated that a high ki67 is associated with a worse disease-free and overall survival in breast cancer. ki67 has been reported as a clinical tool for classification of luminal A and B tumors. In postmenopausal patients with ER-positive tumors who did not receive chemotherapy from the ATAC trial, the prognostic information from IHC4 (ER, PR, HER2, and ki67) was similar to that seen with the 21-gene recurrence score. When IHC4 was compared with the ROR score in this cohort, ROR only improved upon IHC4 in the HER2-negative, node-negative cohort. However, the ideal cutpoint for ki67 remains unclear, likely indicating it should be considered a continuous marker. Furthermore, analytic validity and intraobserver variability in interpreting results remains challenging. These issues continue to affect the clinical utility of ki67 for decision-making for adjuvant therapy.
ki67 has also been explored as an early predictive biomarker for neoadjuvant endocrine therapy. Changes in ki67 can occur early in response to the cytostatic effect of endocrine therapies and have been associated with improved recurrence-free and overall survival. This is being investigated currently in the phase III ALTERNATE trial ( clinicaltrials.gov identifier NCT01953588). This study is enrolling postmenopausal women with clinical stage II–III ER-positive breast cancer who are started on neoadjuvant endocrine therapy. An early biopsy evaluating ki67 is then used to determine whether patients will continue neoadjuvant endocrine therapy or change to neoadjuvant chemotherapy.
Breast Cancer Index (BCI)
BCI is an assay that consists of two gene expression biomarkers: molecular grade index (MGI) and HOXB13/IL17BR. BCI was validated as a prognostic test in blinded retrospective analysis of the Stockholm trial. In this prospective randomized study, node-negative, postmenopausal patients were treated with tamoxifen versus observation. When the BCI was evaluated in 588 ER-positive cases from this parent trial, there was a significant association with distant recurrence and breast cancer death. In tamoxifen-treated patients, cohorts at higher 10-year risk for distant recurrence were identified (risk up to 16.9%; 95% CI 7.2–25.6). However, more than 50% of the patients were able to be identified to have a very low risk of less than 3%, and chemotherapy avoidance may be considered for this low-risk population. A subsequent analysis of tamoxifen-treated breast cancer cohorts demonstrated that BCI also was prognostic for late distant recurrences, occurring after 5 years of tamoxifen. Further support for BCI as a potential predictive biomarker for extended endocrine therapy came from an analysis of MA.17, in which postmenopausal women were randomized to letrozole or placebo after completion of 5 years of adjuvant tamoxifen. A prospective-retrospective case-control study (N = 249) showed that high HOXB13/IL17BR (H/I) ratio was associated with a decrease in recurrence-free survival from extended letrozole therapy (odds ratio = 0.35; 95% CI 0.16–0.75; p = .007). This remained significant when adjusted for clinicopathologic factors. Reduction in absolute risk of recurrence with letrozole was 16.5% in patients with a high H/I ( p = .007). BCI was also compared with IHC4 and the 21-gene recurrence score in the TransATAC cohort from the large randomized phase III ATAC trial where it was found to have prognostic significance for both early (0–5 year) and late (5–10 year) recurrence. Although further validation of BCI is needed, its potential use for both early decision-making regarding chemotherapy and late decisions regarding extended endocrine therapy is appealing.
Summary
The use of these biomarker assays can identify different subsets of HR-positive breast cancers. It is important to develop these tools to identify those patients who have tumors with an excellent prognosis and are sensitive to endocrine therapies and thus avoid the toxicities of chemotherapy in these patients. In addition, it is also critical to identify which patients remain at substantial risk for distant recurrence after 5 years of endocrine therapy for whom longer durations of endocrine therapy are rational. There continues to be development of multiple new assays to answer these clinical questions.
Adjuvant Endocrine Therapy for ER- and/or PR-Positive Breast Cancer
Premenopausal Women
In women under 45 years of age, tamoxifen for 5 years induces an absolute 15-year benefit of 10.6% in overall survival, reducing the 15-year breast cancer mortality from 35.9% to 25.3% (relative risk 0.71; 95% CI 0.61–0.83, p = .00002). The landscape of endocrine therapy for premenopausal breast cancer is in a state of flux, with outstanding questions remaining about the optimal duration, use of ovarian function suppression (OFS), and best agent (tamoxifen versus AI + OFS). Current data and key questions are summarized in this section (see “Duration of Therapy” and “Ovarian Function Suppression”).
Ovarian Function Suppression.
OFS can be achieved medically (via chronic use of luteinizing hormone–releasing hormone [LHRH] agonists), surgically (via bilateral oophorectomy) or via radiation to the ovaries. When considering OFS in premenopausal HR-positive breast cancer, key questions include the following:
- •
What is the benefit (if any) of ovarian function suppression alone compared with tamoxifen alone?
- •
What is the benefit (if any) of ovarian function suppression plus tamoxifen compared with tamoxifen alone in premenopausal women?
- •
What is the best endocrine therapy agent (tamoxifen vs. an AI) to use in premenopausal women who receive ovarian function suppression? And are there other caveats?
- •
Does it matter how ovarian function suppression is achieved: Medically? Surgically? With radiation?
- •
What is the added toxicity of ovarian function suppression?
Table 70.2 summarizes key published trials. Some answers have started to emerge but may need longer follow-up and further confirmation. SOFT (Suppression of Ovarian Function Trial) addressed the added value of ovarian function suppression combined with tamoxifen compared with tamoxifen alone, enrolling 2066 premenopausal women many of whom had also received chemotherapy but remained premenopausal despite this (a confounding factor in prior trials). SOFT results showed that there was no additional benefit in terms of disease-free or overall survival for adding OFS for premenopausal women as a whole. At a median follow-up of 67 months, the 5-year disease-free survival was 86.6% in the tamoxifen plus OFS and 84.7% in the tamoxifen group (hazard ratio 0.83; 95% CI 0.66–1.04; p = .10). However, among the subgroup under age 35 years who had received chemotherapy (only 233 women), the 5-year freedom from breast cancer was 67.7% (95% CI 57.3–76.0) in the tamoxifen alone versus 78.9% (95% CI 69.8–85.5) in the tamoxifen plus OFS group. Small numbers limit the ability to draw firm conclusions. Furthermore, SOFT and E3193 (an underpowered trial examining early-stage breast cancer treated with tamoxifen versus tamoxifen + OFS as the only adjuvant medical therapy) addressed the question of added toxicity over tamoxifen alone and demonstrate an increase in side effects (hot flashes, sexual dysfunction, bone health) resulting in decreased quality of life for the women receiving tamoxifen plus OFS.
| Trial Name | Population (n, Key Inclusion Criteria) | Treatment Arms | Key Outcome |
|---|---|---|---|
| ABCSG-12 | n = 1803, premeno, stage I–II, HR+, <10 + nodes | Tam + Gos 3 y vs. AI + Gos 3 y (also randomized to ZA or not) |
|
| E5188/INT 0101 | N+, HR+, premeno | After CAF, randomized to no treatment vs. Gos 5 y vs. Gos + Tam 5 y |
|
| ZIPP | n = 2076 age < 50 or premeno, Stage I-II | Randomized after primary therapy: no treatment vs. Gos 2y vs. Tam 2 y vs. Gos 2 y + Tam |
|
| E3193 | n = 345 HR+, N−, premeno | Tam 5 y vs. Tam + OFS 5 y |
|
| SOFT | n = 2066, HR+, premeno | Tam 5 y vs. Tam + OFS 5 y (vs. AI + OFS 5 y) |
|
| TEXT/SOFT | n = 4690, premeno | Tam + OFS 5 yr vs. AI + OFS 5 yr |
|
Related posts:
Anatomy of the Breast, Axilla, Chest Wall, and Related Metastatic Sites
Mesenchymal Neoplasms and Primary Lymphomas of the Breast
Therapeutic Value of Axillary Node Dissection and Selective Management of the Axilla in Small Breast Cancers
Surgical Management of Early Breast Cancer
Neoadjuvant Chemotherapy and Radiotherapy
Delayed Diagnosis of Symptomatic Breast Cancer
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
