Although significant advances have been made in the treatment of breast cancer resulting in continued improvement in survival, more than 40,000 women will die of the disease this year.1 Approximately 30% of women diagnosed with early-stage breast cancer will develop a systemic metastatic recurrence, with only a few of these patients achieving long-term survival with standard chemotherapy.2 In addition, 5% to 10% of women are diagnosed with metastatic disease at first presentation of breast cancer. Overall survival (OS) for patients with metastatic disease has changed little over the last 50 years, despite a marked increase in the choice of active agents for treatment. The availability of new targeted biological therapies, particularly trastuzumab for HER-2/neu overexpressing disease, as well as new hormonal and chemotherapeutic agents, has clearly improved outcome for patients with certain biological subtypes of this disease. However, the concept of treating metastatic breast cancer as a chronic disease is still largely theoretical because most patients will survive less than 5 years following diagnosis.

The most significant advances in the treatment of metastatic breast cancer are not limited to the inclusion of new targeted biological therapies. Most recently, the introduction of new cytotoxic agents, new formulations of existing drugs, rationally designed combination therapy, and variation of dose and schedule have resulted in improvements in outcome with generally well-tolerated toxicity profiles. The advantage of these studies is not only seen in improved options for patients with metastatic disease, but also in the ability to test more promising treatment approaches in the adjuvant setting. In addition, there are more options for hormonal therapy, as well as a greater understanding of how and when to use treatment directed toward the estrogen receptor (ER). However, treatments invariably fail, and some tumors are initially resistant to therapy. A better understanding of the biology that drives tumor growth and resistance to therapy has helped to identify new potential targets as well as develop new therapies, but clearly additional study is needed.

The primary goal of treatment for metastatic disease is to control symptoms and to prolong survival in the context of maximizing quality of life. Treatment is palliative, but effective therapy can significantly prolong life. The choice of therapy is based on a variety of factors including biological markers, extent and pattern of disease, prior treatment, patient performance status, and patient preference. Biological markers correlate with the pattern of organ involvement and prognosis; hormone receptor (HR)-positive disease most commonly presents in bone and soft tissues, and visceral involvement is usually limited early in the course of disease. In contrast, visceral involvement dominates in HR-negative and HER-2/neu-positive disease, with risk of pending organ dysfunction. The National Comprehensive Cancer Network provides guidelines for evaluation and treatment of advanced disease that are available online at http://www.nccn.org.

Extent of metastatic disease is an important determinant of therapy and is determined with baseline studies that are also used for future assessment of treatment effect. In addition, organ dysfunction may preclude certain therapies. Therefore, initial evaluation should begin with complete laboratory studies including complete blood count, liver function, renal function, electrolytes, calcium and albumin, as well as scans. In addition to computed tomography (CT) scanning of the chest, abdomen, and pelvis, a bone scan should be obtained. Positron emission tomography/CT scanning is now often used as a single test to evaluate both visceral organs and bone,3 although bone scans may be a more sensitive determinant of osteoblastic lesions.4 One accessible lesion should be biopsied, both to confirm the origin of the metastatic cells5 and to reassess biological markers including ER, progesterone receptor (PR), and HER-2/neu. With progression to metastatic disease, HR status may occasionally change primarily with loss of receptor expression,6 and several reports have suggested that the same is true for HER-2/neu,7 although this remains to be validated. In addition, clinical behavior or a pattern of organ involvement that is inconsistent with marker results should prompt reevaluation because the outcome could have a significant impact on therapeutic decisions.

A variety of factors play a role in determining the most appropriate treatment during the course of metastatic disease. These include biological markers, history and type of adjuvant treatment, disease-free interval, and extent of disease. Cancers are initially divided into 3 major groups: HR positive, HER-2/neu positive, or receptor negative, also referred to as triple negative. Treatment decisions are generally based on these receptors and modified with the factors just listed. Patients with HR-positive disease are usually treated with sequential hormone therapy,8 leaving chemotherapy for the treatment of hormone-resistant disease. Chemotherapy is also used to treat HR-negative disease, and in combination with targeted biological agents including trastuzumab or lapatinib for HER-2/neu positive disease and bevacizumab (see Chapter 88). Table 92-1 outlines chemotherapy options for the treatment of advanced breast cancer, and includes both approved indications as well as regimens for which clinical data exists. Radiographic testing and evaluation of symptoms is used to assess the effectiveness of therapy.9

Hormonal therapy for advanced breast cancer has evolved significantly in the more than 100 years since initial data documenting the effect of ovarian ablation on advanced breast cancer in premenopausal women.10 ER and PR are now routinely measured, and it is now well understood that expression of these receptors determines response to hormone therapy. The approach to treating HR-positive advanced breast cancer must take into account the type and extent of adjuvant hormonal therapy, time since last hormonal treatment, and the biological aggressiveness of the disease.8 Options include sequencing of the nonsteroidal aromatase inhibitors (AI) letrozole and anastrozole, the steroidal AI exemestane, fulvestrant (a novel ER antagonist without agonist effects), tamoxifen (a selective ER modulator), and megestrol acetate (a synthetic progestational agent).11 High doses of estrogen may also be effective; toxicities such as thrombosis, nausea, and bloating may limit therapy.12Table 92-2 lists the hormonal options for the treatment of advanced breast cancer.

It does not appear that the exact sequencing of specific agents is important in terms of outcome. Although in 3 randomized trials AIs were found to be superior or equivalent to tamoxifen as first-line treatment for metastatic disease in terms of response and time to progression (TTP), no survival benefit was found. The recent EFFECT trial randomized almost 700 women with advanced disease recurring or progressing after treatment with a nonsteroidal AI to receive either fulvestrant or exemestane.13 Median TTP was identical at 3.7 months in both groups as was the overall response rate (ORR) and rate of carcinoma of breast (CB). Interestingly, for patients who did respond, the median duration of response was quite long, ranging from 9.8 to 13.5 months, and CB was seen in up to 29% of patients with visceral dominant disease and regardless of prior response to the nonsteroidal AI. Based on these data, it seems that sequential hormonal therapy can continue to provide benefit to patients with hormone responsive disease, despite progression on relatively similar agents. Patients with short response duration in the adjuvant or metastatic setting are less likely to benefit from continued hormonal manipulations.

Sequential Single-Agent Chemotherapy

Given the widespread use of adjuvant chemotherapy, many patients with advanced disease will have received prior treatment with anthracyclines and taxanes. Significant advances in the use of taxanes in the treatment of metastatic disease have improved subsequent options for therapy, as well as the availability of a broad range of active agents.

The taxanes function as inhibitors of the microtubules that are essential for cell division14,15 and are one of the most effective class of agents available for treating both early- and late-stage breast cancer. Paclitaxel and docetaxel, originally identified as natural products from the yew tree, are considered a standard of care in the treatment of metastatic disease16,17 with single- agent response rates (RR) of 32% to 68%.18

Variations in schedule of administration have been tested in both the metastatic and adjuvant settings for both paclitaxel and docetaxel.17,19 Weekly paclitaxel, based on mathematical models by Norton20 that predict superiority for more frequent dosing, appears to offer improved efficacy with reduced hematologic toxicity,21 as well as a potential antiangiogenic effect.22 Cancer and Leukemia Group B (CALGB) 984023 randomized 577 women with taxane-naive metastatic breast cancer to receive weekly paclitaxel versus every 3 week paclitaxel. Weekly paclitaxel was superior to every 3 week paclitaxel with respect to RR (40% vs 28%; p = 0.017) and TTP (9 vs 5 months; p = 0.0008), with less granulocytopenia but more sensory neuropathy from weekly dosing. Other studies evaluating a dose relationship for paclitaxel given every 3 weeks documented improved RRs24 at 175 mg/m2 compared with 135 mg/m2, but no differences in response and greater toxicity when higher doses were used.25 In the treatment of metastatic breast cancer, weekly paclitaxel appears to be superior to every 3 week paclitaxel and is currently the preferred schedule of administration for this taxane.

In contrast to paclitaxel, weekly dosing of docetaxel26 is associated with less bone marrow and neurotoxicity as well as stomatitis but significantly more non-life-threatening toxicities including nail changes (onycholysis) and canalicular stenosis (tearing) compared to every 3 week dosing.27 Thus the standard dosing schedule for docetaxel is every 3 weeks. Increasing every 3 week docetaxel dose to 60, 75, or 100 mg/m2 has been demonstrated to increase ORR27 (22.1%, 23.3%, and 36%, respectively) and TTP, but with a significant increase in both hematologic and nonhematologic toxicity and no benefit in terms of survival.

Head‐to-Head Comparison of Docetaxel to Paclitaxel

Preclinical studies identified potentially important differences between paclitaxel and docetaxel/28 Compared with paclitaxel, docetaxel demonstrated a higher intracellular concentration in target cells29 as well as a greater affinity for the tubulin binding site.30,31 The toxicity of the 2 taxanes differs, with paclitaxel associated more commonly with neurotoxicity and hypersensitivity and docetaxel with fluid retention, asthenia, as well as nail, eye, and skin changes.

An open-label, randomized phase III trial was designed to directly compare paclitaxel and docetaxel as second-line therapy for metastatic breast cancer.32 The 449 patients were assigned to either paclitaxel, 175 mg/m2, or docetaxel, 100 mg/m2, every 3 weeks until progression or toxicity. Although fewer patients discontinued docetaxel for progression than paclitaxel (47% vs 75%), more patients stopped docetaxel for adverse events (26% vs 8%). TTP was prolonged for patients receiving docetaxel (5.7 vs 3.6 months, hazard ratio [HR]: 1.64, p < 0.0001), as was survival, with median overall survival (OS) 15.4 versus 12.7 months (HR: 1.41, p = 0.03). Febrile neutropenia was more frequent in patients receiving docetaxel than paclitaxel (14.9% vs 1.8%), as was asthenia, peripheral edema, stomatitis, and neurosensory changes.

Based on this phase III trial, every 3 week docetaxel appears to be superior in efficacy, but more toxic, than every 3 week paclitaxel. Weekly paclitaxel is superior to every 3 week paclitaxel, bringing into question whether difference in optimal schedule of administration could be responsible for the superiority of docetaxel. To improve quality of life and reduce toxicity, higher doses of docetaxel should be given with prophylactic myeloid growth factors.33

Newer Microtubule Targeted Agents

New formulations of existing agents have the advantage of potentially modifying the toxicity profile as well as possibly improving tumor penetrance leading to enhanced efficacy. Nab-paclitaxel is an albumin-bound, solvent-free novel formulation of the insoluble drug paclitaxel in the form of albumin-based nanoparticles that eliminates the need for premedications and the risk of hypersensitivity from paclitaxel.34 Binding of drug to albumin receptors (glycoprotein 60 [gp60]) on the endothelial cell wall of the tumor neovasculature35 may result in enhanced tumor cell uptake of nab-paclitaxel and therefore enhanced antitumor activity.

A phase III trial in patients with metastatic breast cancer compared nab–paclitaxel (n-pac), 260 mg/m2, to paclitaxel, 175 mg/m2, given every 3 weeks.36 Treatment with n-pac resulted in improved RR (33% vs 19%; p < 0.001) and TTP (5.0 vs 3.7 months; p = 0.030) compared with paclitaxel, with reduced grade 4 neutropenia (9% vs 22%; p < 0.001) despite a 49% higher paclitaxel dose. Grade 3 sensory neuropathy was more common with n-pac than paclitaxel (10% vs 2%; p < 0.001) but improved at a median of 22 days. No hypersensitivity reactions occurred with n-pac despite the absence of premedication and shorter administration time.

Two large phase II studies evaluating weekly n-pac demonstrated RRs of 12% to 15% in taxane-resistant metastatic disease,37,38 with low rates of grade 3 sensory neuropathy and minimal hematologic toxicity. A phase II trial randomized 300 patients with untreated metastatic breast cancer to 4 different arms; A: n-pac, 300 mg/m2 every 3 weeks; B: n-pac, 100 mg/m2 weekly (3 on, 1 off); C: n-pac, 150 mg/m2 weekly (3 on, 1 off); and D: docetaxel, 100 mg/m2 every 3 weeks.39 Weekly n-pac resulted in the best RR (B: 62%, C: 70%, A: 43%, D: 38%), and high dose weekly n-pac was associated with the best progression-free survival (PFS) (arm C vs D, HR: 0.46; p = 0.002; arm C vs B, HR: 0.55; p = 0.009). The least toxicity was seen with arm B.