Treatment of Metastatic Breast Cancer: Endocrine

Stephen R.D. Johnston

Gaia Schiavon

INTRODUCTION

The optimal management of patients with metastatic breast cancer (MBC) remains a challenge. Systemic drug treatments such as chemotherapy, endocrine therapy, biological targeted therapy, and supportive therapies, including bisphosphonates for bone disease, are the mainstay of care. The clinical decision as to which is the most appropriate treatment option is based on a number of patient- and disease-related factors (Table 70-1). Approximately twothirds of human breast carcinomas express estrogen receptors (ER) and thus may be dependent on estrogen for their growth, and for patients in whom their breast cancer (either primary tumor or biopsy of accessible metastatic disease) is positive for ER and/or progesterone receptor (PgR), endocrine therapy is an important treatment option to consider that has minimal toxicity.

For patients with ER/PgR positive (+ve) breast cancer and an estimated low risk of rapid progression of their advanced disease (i.e., soft tissue and/or bone metastases as their dominant site, absence of life-threatening visceral involvement, disease-free interval greater than 2 years, limited sites of metastatic involvement), endocrine therapies can be very effective in the treatment of their advanced/metastatic disease. For example, locally advanced ER +ve disease within the breasts of elderly women is often slow growing and extremely hormone sensitive. Excellent clinical responses can be achieved with simple well-tolerated endocrine therapy such as the antiestrogen tamoxifen, albeit maximal response and tumor shrinkage may take between 6 and 9 months to occur (Fig. 70-1A,B). However, sites of visceral metastases such as the liver may also respond well to endocrine therapy provided appropriate selection of patients is undertaken. For example, postmenopausal patients with strongly ER/PgR+ve disease with a long treatment-free-interval of many years after completion of adjuvant tamoxifen may then develop metastatic disease within the liver but with a limited number of tumors and preserved organ function (i.e., normal liver function tests). They may lack any symptoms from their advanced disease and show good overall performance status. Such patients can have an excellent clinical response to endocrine therapy alone with, for example, aromatase inhibitors (AIs), which may last for 18 to 24 months before their disease progresses and patients require chemotherapy (Fig. 70-1C,D).

Although evidence from randomized trials directly comparing endocrine therapy to chemotherapy as initial first-line treatment for MBC is limited, a Cochrane review found no survival differences but more toxicity associated with initial chemotherapy (1). Therefore, appropriate selection of patients who are suitable for initial endocrine therapy is therefore crucially important in order to maximize the benefits from such treatments, in particular as long-term disease control for up to 18 months with minimal side effects is not uncommon. In this chapter the evidence for each of the current endocrine therapy options that are available for advanced disease in both post- and premenopausal women are reviewed in more detail, together with the emerging strategies that might be used in the future to further enhance their effectiveness. In particular, recent results from several key clinical trials of endocrine therapy (including those in combination with various targeted signaling inhibitors) will be discussed, along with the implications for the optimal sequence of endocrine therapies in advanced breast cancer.

However, one of the key factors that determine whether endocrine therapy will be an effective option for metastatic disease relates to prior exposure to adjuvant endocrine therapy and to the level of hormone receptor expression. These two factors are among the most important in determining a tumor’s underlying endocrine sensitivity, and whether “intrinsic de-novo” or “acquired resistance” will influence the outcome and response to endocrine therapy in the metastatic setting. These issues will be addressed first, before reviewing the clinical data that are available with each of the various current endocrine treatments in first- and secondline settings and the future strategies that might be used to circumvent/prevent endocrine resistance.

TABLE 70-1 Clinical Parameters Utilized in Decision Making for Systemic Therapy Options in Advanced Breast Cancer

Patient Factors

Age

Menopausal status

Performance status

Severity and nature of symptoms

Presence/absence of visceral disease

Organ function (i.e., liver/renal/bone marrow function)

Disease and Treatment-Related Factors

Tumor biology (ER/PgR status; HER2 status)

Dominant site of disease (i.e., bone/soft tissue vs. visceral metastases)

Number of sites of metastases (tumor burden)

Prior adjuvant systemic therapies

Duration of treatment-free period (i.e., sensitive vs. resistant disease)

IMPLICATIONS OF PRIOR ADJUVANT THERAPY AND ER EXPRESSION

Impact of Adjuvant Endocrine Therapy

The use of adjuvant endocrine therapy has significantly improved survival of women diagnosed with ER+ve early stage breast cancer (2). Until recently tamoxifen had been the gold standard of adjuvant endocrine therapy for ER+ve breast cancer. Tamoxifen is a nonsteroidal ER antagonist that inhibits breast cancer growth by competitive antagonism of estrogen at the receptor site. However, its actions are complex due to partial estrogenic agonist effects that in some tissues (i.e., bone) can be beneficial (3) but in others may be harmful, increasing the risk of thrombo-embolism and uterine cancer (4).

The results of the most recent Early Breast Cancer Trialists Collaborative Group overview involving over 21,000 women has shown that tamoxifen for about 5 years reduces the risk of death by around one-third in the first 15 years (RR 0.71) (2). The proportional risk reduction was not significantly affected by age, the use of chemotherapy, nodal status, or expression of PgR, with the absolute benefit relating to absolute risk of recurrence. Despite the significant clinical benefit from tamoxifen in early breast cancer and its impact on improving overall survival (OS), a significant annual hazard rate for recurrence persists in ER+ve breast cancer, with over half the recurrences occurring after 5 years of therapy has completed (2). This raises the issue of whether tamoxifen can be used again in MBC if it has already been given in the adjuvant setting.

In the recent past, AIs have demonstrated superior risk reduction over tamoxifen in postmenopausal early breast cancer. In the ATAC (Arimidex, tamoxifen, Alone or in Combination) trial, anastrozole was compared with tamoxifen and with the combination of the two drugs and was shown to be superior to both in terms of disease-free survival (DFS) (hazard ratio [HR] 0.86, 95% CI, 0.78-0.95; p = .003) in hormone-receptor-positive patients at a median follow up of 120 months (5). Similarly, the BIG 1-98 trial letrozole was significantly better than tamoxifen in terms of both DFS (HR 0.82, 95% CI, 0.74-0.92) and OS (HR 0.79, 95% CI, 0.69-0.90). (6) There have also been several trials that have reported a benefit in risk reduction with the use of adjuvant AIs given after an initial 2 to 3 years of tamoxifen versus tamoxifen for 5 years. While the use of adjuvant AIs in postmenopausal breast cancer has further improved outcomes, for those who still subsequently relapse, this has created new challenges to determine the most appropriate endocrine therapy strategies to utilize in women with metastatic disease.

Initial sensitivity to adjuvant therapy in early breast cancer can be an important predictor for the likelihood of response to further endocrine therapy in the metastatic setting. Some patients relapse very early while taking their adjuvant endocrine therapy. These tumors might be expected to have intrinsic “de-novo” endocrine resistance. Alternatively, some patients may relapse at the end, or shortly after completing adjuvant endocrine therapy, and as such could have developed “acquired” resistance, which may still allow response to alternative endocrine therapies. Alternatively, patients may relapse at a much later time point many years following completion of their adjuvant endocrine therapy (late relapse), and cancer cells in this situation may have retained full endocrine sensitivity. Thus the time point for relapse from diagnosis (disease-free interval) and also from prior adjuvant therapy (treatment-free interval) both might determine the response to further endocrine therapy in the metastatic setting (Fig. 70-2).

Change in Receptor Expression in MBC

It is well recognized that the level of ER expression in breast cancer cells determines the extent of endocrine sensitivity, and therefore the magnitude of benefit that can be obtained from adjuvant endocrine therapy. In the EBCTCG tamoxifen overview, highly ER+ve disease (≥200 fmol/mg as measured by ligand-binding assay) was associated with the greatest benefit with a hazard rate ratio for breast cancer mortality with tamoxifen of 0.53, while those women with the lowest but still measurable ER levels (<9 fmol/mg) did not derive any benefit from adjuvant tamoxifen (2). Utilizing more modern immuno-histochemistry (IHC) assays to measure ER, it was initially thought that tumors where only 1% of cells stained positive for ER by could benefit from adjuvant tamoxifen (7, 8). However, more recent molecular studies have categorized different subtypes of ER+ve breast cancer that may better relate to likely hormone sensitivity and suggest that very few of the ER weak tumors (i.e., those with only 1% to 9% cells ER+ve) show a similar molecular feature to those strongly ER+ve tumors (9). Likewise, tumors that are ER negative (-ve) but express PgR may have a low expectation of clinical benefit from endocrine therapy. As such the primary breast tumor’s level of ER expression (and possibly its molecular subtype) could predict the benefit from endocrine therapy for tumors that recur at loco-regional or distant sites many years later.

Increasingly, it has become important to establish whether the receptor status in the tumor changes during progression from early breast cancer to regional or metastatic recurrence, because this itself may be the most important factor in determining the likelihood of response to further endocrine therapy. Several studies comparing receptor expression between paired biopsies have showed that receptor expression may change, with either a reduction or loss of ER expression in up to 35% of tumors and gain of the human epidermal growth factor receptor 2 (HER2) expression in up to 10% of ER+ve breast cancers (Table 70-2) (10, 11, 12 and 13). Either or both of these phenotypes is known to account for acquired endocrine resistance, but may still allow response to alternative endocrine therapy strategies as illustrated
below due to partial non-cross resistance between different endocrine treatments.

FIGURE 70-1 (A) An elderly woman with a locally advanced ER positive (+ve) slow growing right breast cancer at the time of diagnosis prior starting systemic treatment. (B) Excellent response with significant tumor shrinkage after 6 months of endocrine therapy (tamoxifen). (C and D) Transaxial scan CT images of a patient with limited number of lesions and preserved organ function at baseline (C) and after aromatase inhibitor-based therapy (D) showing partial response.

As such, re-biopsies taken from sites of metastatic disease whenever clinically appropriate and feasible are increasingly recommended in order to plan appropriate systemic further therapy, in particular whether an endocrine approach will be effective option. This has been emphasized in the recent 2012 International Consensus Guidelines for the treatment of advanced disease, recognizing the importance of first- and second-line endocrine therapy as preferential options for patients with potentially endocrine-responsive MBC (14). The options available to these patients are discussed in the next two sections.

FIRST-LINE ENDOCRINE THERAPY OPTIONS FOR POSTMENOPAUSAL MBC

Tamoxifen

Historically, the selective estrogen receptor modifier (SERM) tamoxifen was the standard first-line treatment option for hormone receptor-positive MBC in postmenopausal patients, especially in those presenting with advanced disease or in whom adjuvant endocrine therapy had not been given. In 86 clinical studies involving 5,353 patients, the objective response rate (ORR including complete response [CR] + partial response [PR]) of 34% was observed, with an additional 19% of patients achieving stable disease (SD) for at least 6 months (15). A median duration of response up to 24 months was observed. Response rates (RR) tended to increase with age, with a 27% response in patients younger than 50 years compared with 43% in those over 70. Tamoxifen was generally well tolerated with a low incidence of serious side effects, including a low but significantly increased incidence of endometrial cancer and thromboembolic events due to its partial estrogenic agonist effects (16).

Tamoxifen was compared with many other endocrine therapies (high-dose estrogens, megestrol acetate (MA), oophorectomy, and other SERMs) in randomized trials, and although these trials are small and lacking statistical power by modern standards, tamoxifen was consistently shown to be at least as effective or better, and often with a better toxicity profile (17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 and 30). However, the majority of women with
ER +ve breast cancer have already been treated with tamoxifen in the adjuvant setting, and while in the past tamoxifen therapy was often used again if there was a treatment-free period of several years, nowadays alternative endocrine approaches that deprive tumors of circulating estrogens are utilized in preference as first-line therapy.

FIGURE 70-2 Conventional definitions of endocrine sensitivity/resistance to adjuvant endocrine therapy. The time-point for relapse from diagnosis (disease-free interval) and also from prior adjuvant therapy (treatment-free interval) both might determine the response to further endocrine therapy in the metastatic setting.

Aromatase Inhibitors

From the mid 1990s, the potent third-generation oral AIs become the standard first-line treatment option for postmenopausal patients with ER +ve advanced/metastatic breast cancer. Estrogens are normally synthesized in the ovary in premenopausal women; following menopause, mean plasma estradiol (E2) levels fall from about 400-600 pmol/L to around 25-50 pmol/L. These residual estrogens come solely from peripheral aromatase conversion partic-ularly in subcutaneous fat, and plasma E2 levels correlate with body mass index in postmenopausal women (31). The oral AIs anastrozole (Arimidex™), letrozole (Femara™), and exemestane (Aromasin™) all reduce serum estrogen levels in postmenopausal women by preventing the conversion of adrenal androgens into oestrogens (Fig. 70-3). Anastrozole and letrozole are third-generation nonsteroidal AIs that have similar pharmacokinetics with half-lives of approximately 48 hours, allowing a once-daily schedule (32, 33). Exemestane is a steroidal aromatase inactivator with a half-life of 27 hours (Fig. 70-3) (34). All three compounds are orally active, and based on the clinical trials outlined below (Table 70-3) these drugs were licensed and approved as first-line endocrine treatment for postmenopausal women with ER+ve advanced breast cancer.

The first published data came from two parallel multi-center double-blind randomized controlled trials in which
anastrozole was compared with tamoxifen as first-line therapy in ER+ve breast cancer. The North American study in 353 women showed that anastrozole significantly prolonged the time to disease progression from 5.6 to 11.1 months (p = .005)(35), while in the larger global trial in 668 patients no difference was found between the treatments in terms of median time to progression (TTP) (8.2 months vs. 8.3 months), response rate (RR) (33% both arms), or clinical benefit rate (CBR) (36). The explanation for the different results may have involved a higher proportion of patients with unknown ER status in the second trial, and a subsequent combined analysis of women with just ER+ve disease from both trials confirmed a significant improvement in disease-free survival (8.5 months vs. 7.0 months) in favor of anastrozole (37). Short-term side effects such as hot flashes, vaginal dryness, and headaches were infrequent and similar in both trials in comparison with tamoxifen.

FIGURE 70-3 Chemical structures of some aromatase inhibitors.

TABLE 70-2 Studies Comparing Receptors Status in Primary versus Relapse

Change in receptor status	Amir (n = 280) Prospective Reanalyzed⁽¹⁰⁾	Curigliano (n = 255) Retrospective Liver Only⁽¹¹⁾	Karlsson (n = 470) Retrospective⁽¹²⁾	Lindstrom (n = 104-459) Retrospective⁽¹³⁾
ER+ → ER-	12%	11%	36%	25%
ER- → ER+	13%	26%	22%	8%
HER2+ → HER2-	12%	32%	nr	7%
HER2- → HER2+	5%	6%	nr	5%
ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; n, number; nr, not reported.

TABLE 70-3 Main Randomized Clinical Trials of Different Endocrine Therapies as First-line Treatment in MBC

Study/Arms	n	ORR% (P value)	CBR% (P value)	Median TTP or PFS mo (P value)	Median OS mo (P value)
AI vs. Tamoxifen
Anastrozole vs Tamoxifen⁽³⁵⁾	171	21	59	11.1 (.005)	33
Anastrozole vs Tamoxifen⁽³⁵⁾	182	17	46	5.6	32
Anastrozole vs Tamoxifen⁽³⁶⁾	340	33	56	8.2	38
Anastrozole vs Tamoxifen⁽³⁶⁾	328	33	55	8.3	42
Letrozole vs Tamoxifen^{(38, 39)}	453	32 (.0002)	50 (.0004)	9.4 (<.0001)	34
Letrozole vs Tamoxifen^{(38, 39)}	454	21	38	6.0	30
Exemestane vs Tamoxifen⁽⁴⁰⁾	182	46 (.05)	—	9.9 (.05)	37
Exemestane vs Tamoxifen⁽⁴⁰⁾	189	31	—	5.8	43
Fulvestrant vs. Tamoxifen or AI
Fulvestrant 250 mg monthly vs Tamoxifen⁽⁵¹⁾	313	31.6	54.3	6.8	36.9
Fulvestrant 250 mg monthly vs Tamoxifen⁽⁵¹⁾	274	33.9	62	8.3	38.7
Fulvestrant 250 mg monthly vs Anastrozole⁽⁵²⁾	222	20.7	44.6	5.5	—
Fulvestrant 250 mg monthly vs Anastrozole⁽⁵²⁾	229	15.7	45.0	5.1	—
Fulvestrant 250 mg monthly vs Anastrozole⁽⁵³⁾	206	17.5	42.2	5.4	—
Fulvestrant 250 mg monthly vs Anastrozole⁽⁵³⁾	194	17.5	36.1	3.4	—
Fulvestrant LD + Anastrozole vs Anastrozole⁽⁶¹⁾	258	31.8	55.0	10.8	37.8
Fulvestrant LD + Anastrozole vs Anastrozole⁽⁶¹⁾	256	33.6	55.1	10.2	38.2
Fulvestrant LD + Anastrozole vs Anastrozole⁽⁶⁰⁾	355	—	—	15 (<.007)	47.7 (.049)
Fulvestrant LD + Anastrozole vs Anastrozole⁽⁶⁰⁾	352	—	—	13.5	41.3
Fulvestrant HD vs Anastrozole⁽⁵⁶⁾	102	36.0	72.5	23.4 (.01)	—
Fulvestrant HD vs Anastrozole⁽⁵⁶⁾	103	35.5	67.0	13.1	—
AI, aromatase inhibitors; CBR, clinical benefit rate; HD, high dose (500 mg i.m. at day 0 + 500 mg at days 14 and 28, thereafter 500 mg monthly until progression); LD, loading dose regimen (500 mg on day 0, 250 mg on days 14, 28, and 250 mg every 28 days thereafter); mo, months; n, number; ORR, objective response rate; OS, overall survival; Tam, tamoxifen; PFS, progression-free survival; TTP, time to progression.

The largest single trial was conducted with letrozole in comparison with tamoxifen in over 900 women with advanced breast cancer. Patients treated with letrozole had a significantly higher ORR, CBR, and prolonged TTP (Table 70-3) (38). Of particular note in this trial, nearly 20% patients had received prior tamoxifen in the adjuvant setting, although it had ceased a median of 3 years prior to development of metastatic disease—in this subgroup, retreatment with tamoxifen had a low response rate of 8% compared with a 32% response rate with letrozole. The improvements in clinical efficacy for letrozole resulted in an early improvement in survival during the first 2 years, although with longer followup this difference was lost (39). The explanation for this may relate to the high number (>50%) of patients who prospectively crossed over to the alternate treatment at the time of progression, because significantly more patients benefited from second-line letrozole after progression on tamoxifen than from second-line tamoxifen after letrozole.

Likewise, a European study in 383 patients compared the efficacy and tolerability of the steroidal aromatase inactivator exemestane with tamoxifen as first-line therapy (40). After a median follow-up of 29 months, there was an improvement in progression-free survival (PFS), together with a higher objective response rate with exemestane than tamoxifen (Table 70-3). A subsequent meta-analysis of 6 phase III prospective randomized clinical trials involving 2,787 women treated with second- or third-generation AI versus tamoxifen confirmed a significant advantage in ORR, TTP, and clinical benefit (CB = OR + SD), favoring AIs over tamoxifen (41). However, no difference was found in overall survival. Tamoxifen was associated with a significantly higher incidence of thromboembolic events and vaginal bleeding than the AIs. While some of the smaller trials were not always conducted with the academic rigor of large trials, they all consistently showed small but significant efficacy advantages of the AIs over tamoxifen.

All three drugs (anastrozole, letrozole, exemestane) are thus approved as first-line endocrine therapy options for postmenopausal women with ER+ve advanced breast cancer. It is not clear that one drug is significantly better than any other when direct comparisons have been made, although letrozole achieved greater aromatase inhibition than anastrozole in a crossover pharmaco-dynamic trial (42). Current clinical evidence suggests that there are unlikely to be major direct clinical differences between the different AIs in MBC (43). While there are no comparative data for exemestane with anastrozole or letrozole, further clinical responses have been reported for both exemestane and the second-generation steroidal inhibitor formestane in patients relapsing after anastrozole, letrozole, or the other nonsteroidal inhibitors, suggesting some partial non-cross resistance between the two types of inhibitors (44, 45). In clinical practice, this has meant that exemestane is often used as a second-line option after prior first-line letrozole or anastrozole, as discussed further below.

Fulvestrant (alone or in combination with AI) as First-Line Therapy

Most postmenopausal women with metastatic ER+ve breast cancer have already received either an AI or tamoxifen in the adjuvant setting. Thus, the current clinical challenge is to establish an optimal first-line endocrine therapy for MBC given prior adjuvant endocrine therapy exposure. The ER down-regulator fulvestrant (Faslodex™) is a novel type of antagonist that, unlike tamoxifen, has no known agonist effects (46, 47). Fulvestrant binds to the ER, but due to its steroidal structure and long side-chain, it induces a different conformational shape with the receptor to that achieved by the nonsteroidal antiestrogen tamoxifen. Because of this, fulvestrant prevents ER dimerisation and leads to the rapid degradation of the fulvestrant-ER complex, producing the loss of cellular ER (Fig. 70-4) (48). It has been shown that due to its unique mechanism of action, fulvestrant delays the emergence of acquired resistance compared with tamoxifen in an MCF-7 hormone-sensitive xenograft model (49), and that the lack of agonist effects means fulvestrant does not support the growth of tumors that became resistant to, and subsequently stimulated by, tamoxifen (50).

Early clinical studies showed that fulvestrant at the initially approved dose of 250 mg monthly by intramuscular injection (i.m.) had similar efficacy to tamoxifen as first-line treatment of hormone receptor MBC, with a median TTP of 6.8 and 8.3 months respectively (HR 1.18, 95% CI, 0.98-1.44; p = .088) and ORRs of 31.6% and 33.9% respectively (51). Given the widespread use of prior tamoxifen as adjuvant therapy, two separate phase III first-line trials compared fulvestrant with the AI anastrozole in postmenopausal women with locally advanced or metastatic breast carcinoma who had progressed after prior endocrine therapy (97% with tamoxifen, 56% as adjuvant therapy) (52, 53). These trials were prospectively designed to allow combined analysis of data, and at a median follow-up of 15.1 months fulvestrant was at least as effective as anastrozole in terms of median time to progression (5.5 months vs. 4.1 months, respectively) and objective response (19% vs. 17%, respectively) (54). A subsequent survival analysis after a median follow-up of 27 months showed no significant difference in the median time to death between fulvestrant and anastrozole (27.4 months vs. 27.7 months, respectively) (55).

While these early first-line studies suggested that fulvestrant 250 mg was as effective as either tamoxifen or anastrozole in the first-line setting, more recent first-line studies with fulvestrant have investigated either different loading dose (LD) schedules (LD = 500 mg on day 1, then 250 mg on days 14, 28, and monthly thereafter) or a high dose (HD) schedule (HD = 500 mg on days 1, 14, 28, and monthly thereafter). A phase II trial (FIRST) compared fulvestrant HD with anastrozole as first-line treatment and showed a significantly longer TTP with HD fulvestrant (median TTP not reached versus 12.5 months; HR 0.63; 95% CI, 0.39-1.00; p = .049) (56). As discussed later, a randomized comparison of HD versus LD fulvestrant in the second-line setting (CONFIRM trial) has confirmed the better efficacy for the 500 mg HD schedule (57), resulting in an amendment to fulvestrant’s new drug approval in 2010.

Preclinical evidence from two separate xenograft models suggested that fulvestrant could be significantly more effective when given in a low estrogen environment by combining it with an AI (58, 59). Two randomized phase III trials tested the combination of anastrozole with fulvestrant (LD) versus anastrozole alone as first-line therapy in postmenopausal MBC patients (Table 70-3). SWOG S0226 demonstrated that fulvestrant plus anastrozole significantly improved median PFS (15.0 months vs. 13.5 months) and median OS (47.7 months vs. 41.3 months) compared to anastrozole alone (60). Safety and tolerability were similar between the two treatment arms. Conversely, the FACT trial reported no difference in the median PFS or median OS between the combination and the anastrozole arm (61). The main difference between these two trials was the proportion of endocrine therapy-naïve patients, 60% and 33% in the SWOG and FACT trial, respectively. Indeed, in an unplanned subgroup analysis of the endocrine-naïve subgroup in the SWOG trial, a significantly higher benefit was observed in the combination arm (median PFS 17.0 months versus 12.6 months), not observed in the smaller number of endocrine-naïve patients in the FACT trial. On the basis of this result, a currently active first-line phase III trial (FALCON, NCT01602380) is establishing whether fulvestrant (HD) and anastrozole is better than anastrozole alone in truly endocrine-therapy-naïve patients.

FIGURE 70-4 Diagram summarizing the different “mechanisms of action” of estradiol, tamoxifen, and fulvestrant via the estrogen receptor (ER) leading to different effects in transcription and ultimately tumor cell division.

To date, therefore, the nonsteroidal AIs remain the most effective first-line endocrine option for the majority of postmenopausal patients with MBC. In those with endocrine-sensitive disease, including those who are endocrine-therapy naïve, expected TTP are of the order between 10 to 15 months (Table 70-3). However, the influence of prior adjuvant endocrine therapy remains an important variable in the likelihood of success. While resistance to AIs inevitably develops, it does not preclude further endocrine responses, and effective second-line options are required for these patients.

SECOND-LINE/SEQUENTIAL ENDOCRINE THERAPY FOR POSTMENOPAUSAL MBC

Historically in the 1980s-1990s, tamoxifen was used as a firstline treatment for MBC followed by alternative endocrine therapies in the second-line setting such as the progestin megestrol acetate or the first-generation AI aminoglutethimide. When the third-generation AIs were first investigated as second-line therapy for postmenopausal women progressing after prior tamoxifen, in several RCTs they were shown to be superior in efficacy and/or side effects in this setting (Table 70-4) (62, 63, 64, 65, 66, 67, 68 and 69). However, given their subsequent improved efficacy in the first-line setting versus tamoxifen as discussed above, from 2001 onward the nonsteroidal AIs letrozole or anastrozole became the standard first-line therapy. Thus, it became important to know what the optimal endocrine therapy was in the second-line MBC setting following prior AI therapy, and in particular whether tamoxifen could be used, how effective the partially noncross steroidal aromatase inactivator exemestane was, and what was the role of fulvestrant following prior AI therapy. The evidence for each is discussed below.

Tamoxifen

There are limited trials that have assessed the efficacy of tamoxifen after prior adjuvant AI therapy, and few prospective data show efficacy for tamoxifen where disease had progressed on a nonsteroidal AI (i.e., anastrozole or letrozole). The largest available data come from the letrozole versus tamoxifen study where over 50% of the patients prospectively crossed over to the alternative treatment at the time of progression (39). Median OS from the cross-over data was 19 months for patients who crossed to second-line tamoxifen from their AI, compared with 31 months for patients who crossed to second-line letrozole from tamoxifen. The only other data come from retrospective questionnaire data from the combined analysis of the two international phase III anastrozole versus tamoxifen TARGET trials. This analysis suggested that of 119 patients who went on the receive tamoxifen following progression on anastrozole, 58 patients (49%) derived clinical benefit and 12 patients (10%) had an objective response (70). A subsequent double-blind crossover study by the Swiss centers in the TARGET Trial (SAKK 21/95 sub-trial) further investigated the clinical impact of the sequence anastrozole followed by tamoxifen and reported that 8/16 (50%) derived clinical benefit from tamoxifen (71).

A more contemporary source of data for the efficacy of tamoxifen after prior AI therapy (often in the adjuvant setting) has come from the phase II randomized TAMRAD study testing tamoxifen plus the mTOR inhibitor everolimus (RAD001) (n = 54) versus tamoxifen alone (n = 57) in postmenopausal patients with ER+ve MBC (72). Patients in
the tamoxifen-alone arm had an ORR of 13%, a CBR of 42%, a median TTP of 4.5 months, and a median OS of 32.9 months. These limited data show that tamoxifen may have clinical benefit in almost 50% of patients relapsing on or after an AI, but relatively few obtain an objective response. Further details of this trial are described below.

TABLE 70-4 Main Randomized Clinical Trials of Different Endocrine Therapies as Second-line Treatment or beyond in MBC

Study	Arms	n	ORR% (P value)	Median TTP /PFS mo (P value)	Median OS mo (P value)
AI vs AG and Progestins
Letrozole vs AG⁽⁶⁶⁾	Letrozole 0.5 mg	192	16.7	3.3	21 (.04)^a
	Letrozole 2.5 mg	185	19.5	3.4 (.008)^a	28 (.002)^a
	AG 250 mg × 2	178	12.4	3.2	20
Letrozole vs MA⁽⁶⁵⁾	Letrozole 0.5 mg	188	12.8	5.1 (.02)	21.05 (.03)^a
	Letrozole 2.5 mg	174	23.6 (.04)^a	5.6 (.07)	25.3
	MA 160 mg	189	16.4	5.5	21.5
Letrozole vs MA⁽⁶⁹⁾	Letrozole 0.5 mg	202	21	6.0 (.044)^a	33
	Letrozole 2.5 mg	199	16	3.0	29
	MA 40 mg × 4	201	15	3.0	26
Anastrozole vs MA⁽⁶⁴⁾	Anastrozole 1 mg	263	12.5	4.8	26.7 (.025)^a
	Anastrozole 10 mg	248	12.5	5.3	25.5
	MA 40 mg × 4	253	12.3	4.6	22.5
Exemestane vs MA⁽⁶⁸⁾	Exemestane 25 mg	366	15	4.7 (.037)^a	– (.039)^a
	MA 40 mg × 4	403	12.4	3.9	28.8
Fulvestrant Trials
CONFIRM^{(57, 84)}	Fulvestrant HD	362	9.1	6.5 (.006)	25.2
(2^nd line) Phase III	Fulvestrant 250 mg monthly	374	10.2	5.5	22.8
EFECT⁽⁸¹⁾	Fulvestrant LD	351	7.4	3.7	nr
(3rd line or more) Phase III	Exemestane	342	6.7	3.7	nr
SOFEA⁽⁸²⁾ (acquired AI resistance) Phase III	Fulvestrant	243	7.4	4.4	20.2
SOFEA⁽⁸²⁾ (acquired AI resistance) Phase III	LD+Anastrozole	231	6.9	4.8	19.4
	Fulvestrant LD Exemestane	249	3.6	3.4	21.6
AG, aminoglutethimide; AI, aromatase inhibitor; HD, high dose (500 mg i.m. at day 0 + 500 mg at days 14 and 28, thereafter 500 mg monthly until progression); LD, loading dose regimen (500 mg on day 0, 250 mg on days 14, 28, and 250 mg every 28 days thereafter); MA, megestrol acetate; mo, months; n, number; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; TTP, time to progression. ^aP vs MA (or AG).

Thus, tamoxifen may have some limited efficacy as second-line therapy after an AI although data are sparse to confidently determine the optimal sequence. However, preclinical studies (discussed later) suggest that tamoxifen may be an agonist in cells resistance to long-term estrogen deprivation, and that more effective endocrine/signaling strategies may exist for use following failure of first-line aromatase inhibitor therapy.

Exemestane

Steroidal AIs such as exemestane have an androgen structure and compete with the aromatase substrate androstenedione. They inactivate aromatase by irreversibly binding to its catalytic site, and additional aromatase must be produced before estrogen biosynthesis can resume. Early data suggested a lack of cross-resistance between steroidal AIs and nonsteroidal AIs and that steroidal AIs may be an option in nonsteroidal AI-resistant disease (73). In a phase II, open-label, multinational trial, 24% of patients overall achieved clinical benefit with exemestane following either aminoglutethimide (n = 136) or nonsteroidal AI treatment (n = 105) (45). The ORR and CBR were 8% and 27%, respectively, for patients who received prior aminoglutethimide, with 5% and 20%, respectively, for those who had previously received nonsteroidal AIs.

A separate retrospective analysis of 96 patients receiving exemestane, most of whom had received prior nonsteroidal AIs, reported that 39% experienced clinical benefit with exemestane (74). Lack of cross-resistance between exemestane and nonsteroidal AIs (NSAIs) was also reported in an open-label, exploratory trial comparing sequential treatment with exemestane and NSAIs in MBC (75). Exemestane showed activity in patients after relapse or lack of response to letrozole/anastrozole with an ORR of 8.7%, a CBR of 43.5%, a median TTP of 5.1 months, and a median OS 27.2 of months. As such, exemestane became a standard secondline treatment option for postmenopausal MBC after failure of first-line nonsteroidal AI therapy. As discussed below, exemestane was chosen as the control arm for two large phase III trials that assessed the steroidal antiestrogen fulvestrant in patients no longer responding to a NSAI.

Fulvestrant

Early clinical data with fulvestrant in advanced breast cancer following resistance to AIs came from several phase II studies that showed that fulvestrant 250 mg monthly produced clinical benefit in 20% to 52% patients who had received, and had progressed on, prior treatment with tamoxifen and a nonsteroidal AI (76, 77, 78, 79 and 80). These results suggested that in addition to being effective after prior tamoxifen, disease progression after nonsteroidal AIs may not preclude subsequent treatment with fulvestrant. Two phase III studies have compared fulvestrant with exemestane, which is a recognized standard of care in this setting (Table 70-4).

Only gold members can continue reading. Log In or Register to continue