Luminal HER2-negative (ER/PgR+/HER2-) BC in 2010 in the USA accounted for 61.2% of primary stage IV BC and, as it constitutes almost three quarters of all newly diagnosed BC, remains also the most prevalent subtype among patients who relapse following treatment for early disease [7].

The systemic treatment of choice in luminal BC is ET; it should be considered in all patients, unless contraindications exist, as its use is associated with better health-related quality of life (QoL), greater satisfaction with treatment, less treatment-related side effects and less activity impairment [8]. Contraindications to ET include massive visceral involvement/directly life-threatening disease (visceral crisis) and proven resistance to ET. Unfortunately data supporting the informed choice between ET and ChT is rather limited and comes mainly from non-randomized comparisons. All available randomized studies were conducted in the 1970s and 1980s and compared often obsolete ChT schedules and suboptimal endocrine agents. A meta-analysis of these studies demonstrated, however, no effect of the treatment strategy on overall survival (OS) in spite of a higher response rate (RR) in the ChT group [9]. No randomized comparisons are available between aromatase inhibitors (AI) or fulvestrant and modern chemotherapeutic agents, such as taxanes, capecitabine, vinorelbine or eribulin. Lack of high level of evidence supporting treatment choices in metastatic luminal BC forces oncologists to rely on indirect evidence from retrospective studies or prospective non-comparative data. Although fair comparisons of results in patients treated with ChT and ET outside studies randomizing subjects between these two options are impossible, in general, those treated with first-line ET achieve longer progression-free survival (PFS) and OS. Obviously, populations selected for ChT and ET are different, but these differences are smaller than could be expected: in general, the percentage of ER/PgR-positive patients in ChT studies ranges between 70% and 80%, whereas visceral involvement is present in about 50–80% of patients undergoing ChT and in about 50% of those treated with ET [10].

The main ET options include selective oestrogen receptor modulators (SERM, tamoxifen), selective oestrogen receptor downregulators (SERD, fulvestrant) and AIs. Premenopausal patients should additionally undergo oophorectomy or medical castration by use of luteinizing hormone-releasing hormone (LHRH) agonists. In selected cases with a good response to previous lines of ET, more toxic compounds such as progestins, oestrogens or androgens can also be considered. Available clinical data don’t point to obvious superiority of any of the available ET compounds. Some of first-line studies comparing tamoxifen with AI have reported improvements in response rates (RR) and time to progression (TTP) for AIs. In the second-line setting, AIs have consistently demonstrated improved efficacy and decreased toxicity in comparison to the first-generation AI aminoglutethimide and megestrol acetate [11]. Additionally, a minor OS benefit (HR 0.9) of AI vs. non-AI treatment was seen in the Cochrane review of available studies [11]. Early studies with low-dose (250 mg every 4 weeks) fulvestrant showed no additional benefit from this agent [12]; more promising are the data for high-dose fulvestrant (500 mg 4 weekly + an additional dose after the first 2 weeks). A greater degree of efficacy, both in terms of TTP and OS, of high-dose fulvestrant was first demonstrated in the CONFIRM study comparing it with the originally approved dose of 250 mg [13]. Particularly, promising are the data from the phase II FIRST study comparing high-dose fulvestrant to anastrozole in first-line treatment, which demonstrated significant and clinically relevant OS prolongation (HR 0.7) [14], and results of the confirmatory phase III FALCON study have recently been released demonstrating significant PFS prolongation [15]. In a single study (SWOG S0226), the combined use of fulvestrant (low dose) and an AI vs. an AI alone resulted in significant PFS and OS prolongation, with the largest benefit in patients not exposed to prior tamoxifen and in those with the longest DFI [16]; this was not, however, confirmed in the subsequent study (FACT) and in the meta-analysis of both [17, 18].

In premenopausal patients, the only compound with demonstrated activity as monotherapy is tamoxifen; however, available randomized trials demonstrated RR, PFS and OS improvement from the addition of a LHRH agonist to tamoxifen, so combination treatment is recommended [19]. AI are inactive as single agents in premenopausal patients and should always be combined with ovarian ablation or suppression. Caution is needed to assure «postmenopausal» oestrogen levels in AI-treated patients, as in approximately 40% of those who developed amenorrhoea as a result of ChT and in 17–25% of patients on an AI plus LHRH agonist therapeutically effective levels of oestrogen suppression may not be present [20, 21]. Ovarian suppression or ablation should be continued throughout the course of ET [22]. No data are available on the efficacy of fulvestrant as a single agent in premenopausal MBC women, although its mechanism of action and a single preoperative study suggests it may actually be effective [23].

Particular ET compounds differ in their toxicity profiles: AI use is related to an increased risk of arthralgia, myalgia, tendonitis and carpal tunnel syndrome, vaginal dryness and decreased libido, hot flashes, bone mineral loss and fractures, and tamoxifen use increases the risk of menopausal symptoms including hot flashes and gynaecologic complications as well as endometrial hyperplasia and cancer, and thromboembolic events [24, 25]. Fulvestrant in comparison to «other endocrine therapy» (mostly AI) demonstrated a similar toxicity profile, except for a lower incidence of arthralgia [12].

The general philosophy of systemic therapy for metastatic luminal breast cancer is to continue treatment with endocrine agents for as long as the patient is deriving benefit [3, 4, 26] which means that patients progressing following effective first-line ET should be offered next ET lines, and switching to ChT should be undertaken only in those with proven resistance to endocrine manoeuvres or with rapidly progressive visceral disease (◘ Fig. 49.3). The optimal ET sequence is unknown, and treatment choice depends on menopausal status, prior ET, response duration, drug toxicity profile and availability and the patient’s preferences.

Importantly, among ET-treated luminal MBC, the tumour response is not a surrogate for long-term benefit, and similar OS is observed in those achieving objective tumour regression or long-term disease stabilization [27]. It thus needs to be kept in mind that, as most patients undergoing early lines of treatment for MBC are asymptomatic or mildly symptomatic and do not require rapid symptomatic improvement, the aim of the treatment in the majority of cases is delaying disease progression (prolonging PFS) and not necessarily obtaining tumour shrinkage.

Additionally, not all BCs are equally sensitive to ChT, and in some patients, ET may actually be more effective. As demonstrated in neoadjuvant studies, the frequency of complete pathological remissions following ChT varies significantly between BC phenotypes, being the lowest for ER/PgR+/HER2- patients, in particular for lobular cancers [28–30]. Additionally, some of the effect of ChT in oestrogen receptor (ER)-positive premenopausal patients may not necessary come from its cytotoxic-antineoplastic activity. As demonstrated in the adjuvant NSABP B-30 and IBCSG 13–93 studies, long-term outcomes are superior in patients who become amenorrhoeic as a result of ChT [31, 32], so it can be hypothesized that at least some of the therapeutic effect in this population is actually effected by the endocrine mechanism of action and could be more simply obtained directly with ET.

ET is also feasible in patients with visceral involvement, as long as there is no directly life-threatening disease (visceral crisis) [3, 4, 26]. Indeed, as demonstrated in data from 1396 patients from four phase III studies of first-line ET, the response rate is higher in non-visceral metastases, but if disease control is achieved, its duration is equal in patient with and without visceral involvement [33].

Unfortunately, no biomarkers predictive of ET benefit beyond ER and PgR have been identified for luminal HER2-negative BC. Some help may be provided by analysis of clinical factors, although results are often inconsistent. Various studies suggest the largest benefit from ET is seen in patients with lower tumour grade; longer disease-free interval; lack of liver, CNS or multiple-site involvement; no history of adjuvant ET; strong expression of ER and PgR; a history of clinical benefit from first-line ET; and a lower tumour proliferation rate (identified by Ki67 expression) [34–39]. A number of novel molecular factors potentially predictive of endocrine responsiveness have been described, but none has been properly validated. Recently, the prognostic role of the 21-gene recurrence score (Oncotype DX) for both TTP and 2-year OS in ER/PgR+/HER2- patients was demonstrated in a prospective series of de novo stage IV BC [40].

ET (and indeed any other therapy) in MBC has limited activity, and all patients inevitably develop progressive disease. This endocrine resistance in MBC, for the purpose of patient stratification for clinical research, has been somewhat artificially divided into primary resistance, defined as progressive disease (PD) within the first 6 months of first-line ET for MBC, while on ET and secondary resistance, i.e. PD developing 6 or more months after initiating ET for MBC, while on ET [3, 4]. There are various mechanisms suggested as responsible for endocrine resistance development, including ER loss, ER mutation or activation of alternative signaling pathways.

Loss of the oestrogen receptor may be caused by epigenetic changes to the ESR1 gene coding for the ER (methylation, histone modifications), and studies are ongoing testing histone deacetylase inhibitors (HDACi, which inhibit a key means of epigenetic regulation of gene expression) in combination with ET. A phase II ENCORE 201 study evaluating the combination of exemestane and HDACi entinostat demonstrated OS prolongation in patients progressing on a non-steroidal AI, and the compound was granted «breakthrough therapy designation» by the FDA; confirmatory phase III study results are awaited [41].

Genomic alterations in the ER, very rare in early breast cancer, are observed with increasing frequency in advanced disease, reaching >30% in late MBC [42, 43]. They usually include gain of function point mutations in the ligand-binding domain of the ER protein leading to its constitutive (ligand-independent) activity [44].

Development of a malignant phenotype is associated with dysregulation of numerous cell signaling pathways. Those implicated in the development of endocrine resistance and potentially pharmacologically «targetable» include activation of tyrosine kinase growth factor receptors, PI3K-Akt-mTOR pathway and dysregulation of the cell cycle [45].

Targeting growth factor receptors has provided the most valuable clinical data for HER2-positive disease; this is further discussed in the «Luminal B HER2-Positive Breast Cancer» section. Studies evaluating the role of blocking other growth factor receptors (EGFR, HER3, IGF, FGFR, MET) in combination with ET have so far not been successful [45].

Alterations of the PI3K/AKT/mTOR pathway are among the most frequent genomic abnormalities present in luminal BC [46]. In ER-positive breast cancer cell lines and tumour models, PI3K pathway activation confers endocrine resistance via crosstalk between ER and the receptor tyrosine kinases activating the pathway, as well as through ligand-independent ER activation through mTORC1 [47].

The first approved treatment targeting the PI3K/AKT/mTOR pathway in combination with ET was the mTOR inhibitor everolimus combined with exemestane. In the randomized phase III BOLERO-2 study, clinically significant PFS prolongation has been demonstrated for this combination in patients progressing on or after treatment with a non-steroidal AI; unfortunately this has not translated into an OS benefit, and the treatment was associated with significant toxicities, often leading to treatment interruptions and dose reductions, occasionally requiring permanent drug discontinuation [48, 49]. Phase II experience with PI3K inhibitors both in combination with ET and ChT has so far been disappointing, at least partially due to toxicity limiting optimal dosing [50, 51]. In the phase III BELLE-2 study, PFS prolongation was seen in patients treated with a combination of a pan-PI3K inhibitor buparlisib and fulvestrant, with the largest benefit observed in patients with PIK3CA mutations present in circulating tumour DNA, possibly providing a predictive factor for this therapy [52]. Unfortunately, significant toxicity (transaminase rise, rash, hyperglycaemia and mood disorders) led to frequent treatment discontinuations, potentially limiting treatment applicability.

Other attractive targets are cyclin-dependent kinases (CDK) which control the cell cycle. Combination of the CDK4/CDK6 inhibitor palbociclib and ET in patients with advanced ER/PgR+/HER2- BC resulted in a significant PFS prolongation in three randomized trials either in combination with an aromatase inhibitor in the first-line setting (PALOMA-1, PALOMA-2) or with fulvestrant in patients who had relapsed or progressed during prior endocrine therapy (PALOMA-3) [53, 54]. This has led to accelerated FDA approval of this compound in patients with ER-positive advanced BC as initial endocrine-based therapy for metastatic disease and recently also for second-line therapy. Unfortunately, no OS benefit has been so far observed [53]. Recently, positive results of a phase III MONALEESA-2 study of letrozole ± ribociclib and MONARCH 2 study (of fulvestrant +/– abemaciclib) (another CDK4/CDK6 inhibitor) were also announced, and drug approval [55, 56]. The advantage of agents in this class is their favourable toxicity profile.

As an interaction exists between endocrine regulation and angiogenesis mediated by VEGF, and higher levels of VEGF are associated with a decreased response to ET [57, 58], and the combination of anti-VEGF treatment with ET could potentially increase the efficacy of ET. To test this hypothesis, bevacizumab (a monoclonal antibody against VEGF) was evaluated in combination with endocrine treatment in the LEA and CALGB 40503 trials. Although the CALGB 40503 trial demonstrated a PFS improvement for the combination of letrozole and bevacizumab, neither of the studies demonstrated an OS benefit, and the combined treatment was associated with increased toxicity [59, 60].

In spite of the endocrine responsiveness and the preference for ET, all patients at some stage require cytotoxic ChT. Its principles, however, do not differ from those in other BC subtypes and will be discussed in the «triple-negative BC» section.

Human epidermal growth factor receptor type 2 (HER2) is overexpressed in approximately 15 to 20% of invasive breast cancers and was historically associated with a poor prognosis [61]. The development of HER2-directed therapies in the past 15 years has changed the natural course of the disease and offers patients with HER2-positive BC survival at least equivalent to patients without HER2 alteration. Currently, median OS in metastatic HER2-positive BC exceeds 50 months, which compares favourably with survival of approximately 20 months observed before the anti-HER2 treatment era [2]. Approved HER2-directed agents include trastuzumab (human monoclonal IgG antibody that selectively targets HER2), pertuzumab (monoclonal antibody that blocks the formation of HER2:HER3 heterodimers), trastuzumab emtansine (conjugate of trastuzumab and cytotoxic emtansine – T-DM1) and lapatinib (an oral tyrosine kinase inhibitor that functions downstream of HER2, inhibiting both EGFR and HER2 receptors).

There are several rules guiding the treatment of HER2-positive metastatic disease:

Due to the risk of cardiac toxicity associated predominantly with trastuzumab, cardiac assessment is recommended before initiation of treatment and regularly during therapy. Serious cardiac comorbidities and/or a left ventricular ejection fraction of less than 50% are contraindication to the treatment. In addition, trastuzumab should not be administered concurrently with anthracyclines.

HER2-positive BC represents two molecular subtypes: HER2-positive non-luminal and HER2-positive luminal BC (co-expressing HER2 and ER/PgR). In both groups, anti-HER2 agents are important components of therapy; however, there are substantial differences regarding tumour biology and treatment choices.

Despite the breadth of trials with HER2-directed agents, a number of questions remain regarding the optimal use of anti-HER2-targeted therapy in MBC such as treatment sequence, regimens, duration of the ChT component and the combination with ET in luminal HER2-positive tumours.

Metastatic triple-negative breast cancer (TNBC) compared to other BC subtypes is characterized by aggressive biology, significantly shorter disease-free and overall survival times and a tendency toward visceral (vs. bone) metastases. Patients with early-stage TNBC have a high risk for early relapse. The pattern of relapse in TNBC is distinct, with a rapidly rising rate in the first 2 years following diagnosis, reaching a peak at 2 to 3 years and a decline in recurrence risk over the next 5 years [81]. Unlike other subtypes, no targeted therapy has proven efficacy in the treatment of TNBC; thus, chemotherapy (ChT) remains a mainstay of systemic treatment. ChT may be used also in the treatment of other BC subtypes, i.e. in endocrine-resistant luminal BC or in patients with rapid progression, symptomatic or high-volume visceral disease, in whom ET is unlikely to result in prompt response and clinical improvement. Most of the general principles regarding ChT discussed in this chapter apply also to advanced breast cancers of other phenotypes requiring ChT.