There are a number of agents with activity in metastatic breast cancer. Because the taxanes and anthracyclines are most commonly administered, especially in the first-line treatment of metastatic breast cancer, they are presented first.
Taxanes are among the most active agents for metastatic breast cancer. Agents in this class include:
- Docetaxel – Docetaxel can be administered every three weeks (80 to 100 mg/m2) or weekly (30 to 40 mg/m2 weekly for three weeks followed by one week off) . Of these schedules, we prefer dosing every three weeks based on the results of a randomized trial in the adjuvant setting that showed every three-week dosing results in an improvement in disease-free survival (DFS) compared with weekly dosing . Docetaxel is associated with a significant risk of fluid retention, which is reduced by premedication with dexamethasone .
- Paclitaxel – Paclitaxel can be administered weekly (80 to 100 mg/m2 on days 1, 8, and 15 of a 28-day cycle) or every three weeks (175 mg/m2) [12,13,15]. Whenever possible, we prefer weekly scheduling based on the results of a 2010 meta-analysis, which showed that compared with every three-week treatment, weekly administration of paclitaxel resulted in an improvement in overall survival (OS, hazard ratio [HR] 0.78, 95% CI 0.67-0.89) .
It should be noted that patients treated with paclitaxel are at risk for allergic reactions as a result of the composition of paclitaxel, which is mixed with Cremophor. At most institutions, steroid premedication (dexamethasone 20 mg the night before and morning of infusion) is administered, although it can usually be discontinued if the first two or three doses are tolerated. However, the schedule and administration of dexamethasone varies by institution.
- Nab-paclitaxel – Nab-paclitaxel has activity in metastatic breast cancer similar to other taxanes [16-18]. It may be of particular benefit to patients who are at risk for hyperglycemia and those who cannot tolerate steroids. Nab-paclitaxel has a lower risk of allergic reactions compared with other taxanes, which negates the requirement for steroid premedications and the risk of steroid-induced hyperglycemia.
For patients in whom a taxane is indicated, the choice between taxanes can be based on their comparative safety profiles and patient preferences regarding scheduling of treatments. For example:
- The risks of neuropathy and myalgia are greater with paclitaxel than with docetaxel.
- Paclitaxel can be administered in the setting of mild-moderate hepatic dysfunction. In contrast, docetaxel should not be administered in this context.
- Docetaxel given every three weeks is the more myelosuppressive taxane agent. Risks from docetaxel also include febrile neutropenia, edema, and gastrointestinal toxicities.
There are limited data comparing each of the taxanes against each other. However, they show that the activity and toxicity differ by which schedule was used (ie, weekly or every three weeks) and by agent. As examples:
- Docetaxel was compared with paclitaxel (both on a 21-day cycle) in a trial of 449 patients with advanced breast cancer that had progressed after an anthracycline-containing chemotherapy regimen. Docetaxel produced a significantly better median time to progression (TTP, 5.7 versus 3.6 months) and OS (15.4 versus 12.7 months) compared with paclitaxel . However, both hematologic and nonhematologic toxicity were worse with docetaxel.
Although this study found that every three-week dosing of docetaxel is superior to the same schedule using paclitaxel, weekly paclitaxel (which is the preferred method of administration) has not been compared with every three-week docetaxel in the metastatic setting.
- Paclitaxel and nab-paclitaxel were evaluated as a first-line treatment (as single agents or with optional administration with bevacizumab) in the Alliance trial, conducted by Cancer and Leukemia Group B (CALGB) and the North Central Clinical Trials Group (NCCTG) (CALGB 40502/NCCTG N063H) . The study randomized 799 patients (44% who were previously treated with adjuvant paclitaxel) to bevacizumab with either weekly treatment with paclitaxel (90 mg/m2) or nab-paclitaxel (150 mg/m2) on a three week on, one week off schedule. A third arm including weekly ixabepilone (16 mg/m2) was closed for futility at the first interim analysis. The results for paclitaxel versus nab-paclitaxel are discussed below.
- No significant difference in progression-free survival between paclitaxel and nab-paclitaxel (11 versus 9.3 months, respectively; HR 1.20; 95% CI 1.00-1.40).
- No significant difference in overall survival between paclitaxel and nab-paclitaxel (26.5 versus 23.5 months, respectively, HR 1.17, 95% CI 0.92-1.47).
- A higher rate of serious toxicity (grade 3 or higher) in the nab-paclitaxel versus paclitaxel arms, including sensory neuropathy (27 versus 18%, respectively) and hematologic toxicity (55 versus 22%).
The anthracyclines are important agents for the treatment of breast cancer. However, their use in the adjuvant context often limits their application in women with metastatic disease. Despite this, anthracyclines may be appropriate in select patients, particularly those who are chemotherapy naive and those who were not treated with an anthracycline in the past. In addition, anthracyclines can be used in patients with mild to moderate hepatic dysfunction with dose modification.
The anthracyclines used in the treatment of metastatic breast cancer are:
- Doxorubicin (60 to 75 mg/m2 every three weeks, or 20 mg/m2 weekly for three weeks followed by one week off) – ORR 30 to 47% [21,22]
- Epirubicin (75 to 100 mg/m2 every three weeks, or 20 to 30 mg/m2 weekly for three weeks followed by one week off) – ORR 42 to 50% [23-25]
- Pegylated liposomal doxorubicin (40 mg/m2 every four weeks) – ORR 10 to 33% [26,27]
One potential downside of using anthracycline regimens is the risk for cumulative cardiac toxicity, which may limit the duration of anthracycline-based therapy. However, for patients who are responding to treatment and otherwise are tolerating therapy, the use of dexrazoxane may minimize the risk of treatment-related cardiac damage. For patients treated with doxorubicin, dexrazoxane is indicated after a cumulative doxorubicin dose of 300 mg/m2.
Our preferred anthracycline is doxorubicin or epirubicin because they are both relatively easy to administer. A choice between them is based on geographic and institutional preferences. For example, doxorubicin is more commonly used in the United States while epirubicin is more commonly used in Europe. Although there are no prospective trials comparing it to standard dosing every three weeks, we prefer the use of weekly anthracycline dosing in the metastatic setting because it is better tolerated.
For patients who desire a less frequent administration schedule, pegylated liposomal doxorubicin administered every four weeks appears to be equally active and less toxic compared with doxorubicin administered every three weeks. This was shown in a trial of 509 patients with metastatic breast cancer (56% who had previously received anthracyclines) who were randomly assigned treatment with pegylated liposomal doxorubicin 50 mg/m2 given every four weeks or doxorubicin 60 mg/m2 given every three weeks . Compared with pegylated liposomal doxorubicin, doxorubicin resulted in:
- Slightly higher ORR (38 versus 33%)
- Similar PFS (median, 7.8 versus 6.9 months; HR 1.0, 95% CI 0.82-1.22) and OS (median, 22 versus 21 months; HR, 0.94, 95% CI 0.74-1.19), though PFS measures were confounded by more frequent assessments in the every three-week doxorubicin arm.
- An increase in the risk cardiotoxicity (26 versus 7%, HR 3.16; 95% CI 1.58-6.31).
- Higher rates of alopecia (66 versus 20%), nausea (53 versus 37%), vomiting (31 versus 19%), and neutropenia (10 versus 4). In contrast, pegylated liposomal doxorubicin was associated with a higher rate of plantar-plantar erythrodysesthesia (48 versus 2%), stomatitis (22 versus 15%), and mucositis (23 versus 13%).
Anthracycline versus taxane
There is no evidence of superiority of either anthracyclines or taxanes in the metastatic setting, although the duration of treatment using anthracyclines is more likely to be limited due to the cumulative risk of cardiac toxicity.
While a 2008 meta-analysis of individual patient data (n = 919 patients) found that administration of an anthracycline resulted in an improvement in the ORR (38 versus 33%) and PFS (median, 7 versus 5 months) compared with taxanes , we do not feel that these small differences in ORR and PFS are clinically significant in the current era in which multiple other therapies are available. In addition, the analysis was limited by multiple factors including:
- Heterogeneity between the included trials
- Differences in administration schedules for the taxane used (including the lack of inclusion of trials using weekly administration of paclitaxel)
- The inclusion of patients with or without prior exposure to endocrine therapy
- Lack of inclusion of patients treated with adjuvant taxanes
In our practice, single-agent capecitabine (1000 to 1250 mg/m2 twice daily for 14 days followed by seven days of rest) is a frequent choice as a first-line treatment for metastatic breast cancer, particularly in patients with bone-predominant, estrogen receptor-positive metastatic disease who have progressed despite at least two trials of endocrine therapy. In addition, capecitabine also appears to cross the blood brain barrier better than some agents and may be a good consideration in patients with a history of central nervous system metastases .
Capecitabine is a prodrug of the anti-metabolite fluorouracil. It is orally available, and unlike many agents used in the treatment of breast cancer, it causes very little alopecia or neuropathy. Its primary toxicities are hand-foot syndrome and diarrhea, and it can be used in settings of hepatic dysfunction. The benefit of capecitabine was shown in two multicenter single-arm phase II trials [30,31]:
- In one study, 126 patients were treated with capecitabine (1250 mg/m2 dose). The median TTP was 5 months and the ORR was 28%. Median OS was 15 months .
- In a second study, 95 women were randomly assigned to capecitabine or cyclophosphamide, methotrexate, plus fluorouracil (CMF). Capecitabine resulted in a higher ORR compared with CMF (30 versus 16%, respectively). The median TTP was similar (4 versus 3 months) but capecitabine resulted in a slightly longer median OS (20 versus 17 months) .
For patients who are not candidates for the agents above, others are available and have documented activity against breast cancer.
Eribulin mesylate (1.4 mg/m2 days 1 and 8 every 21 days) is derived from a marine sponge and inhibits the polymerization of tubulin and microtubules. It results in less neuropathy than other microtubule-directed agents and can be administered with dose adjustment for mild to moderate hepatic dysfunction. Therefore, it is a good agent to administer in these situations.
The activity of eribulin was shown in a phase III trial of 762 heavily pretreated patients who were randomly assigned to treatment with eribulin or other chemotherapy (based on physician’s and patient’s choice) . Treatment with eribulin significantly improved OS (median, 13.1 versus 10.6 months). The primary toxicity with eribulin was neutropenia, with grade 3 and 4 neutropenia in 45% of patients, and grade 3 and 4 febrile neutropenia in 5%. Peripheral neuropathy was the most common adverse event leading to discontinuation of eribulin, occurring in 5% of patients.
Of note, a subsequent randomized trial was performed in women with metastatic breast cancer who had received prior anthracycline and taxane therapy with an aim to formerly evaluate eribulin versus capecitabine as first, second, or third-line therapy . Unlike the trial above, there was no difference between eribulin and capecitabine in terms of PFS (four months in each) or overall response rates (11 and 11.5%, respectively). In addition, there was no clinically meaningful difference in OS (15.9 versus 14.5 months, respectively; HR 0.88, 95% CI 0.77-1.00).
Vinorelbine is an intravenously administered agent usually dosed at 30 mg/m2 on a weekly schedule (days 1 and 8 every 21 days) . Vinorelbine causes little nausea, vomiting, and hair loss, and is active as a single agent (ORR 25 to 45%), even in heavily pretreated patients [35-37].
Although data suggest gemcitabine is active in combination with paclitaxel in first-line metastatic breast cancer, gemcitabine (commonly 1000 mg/m2 days 1 and 8 of a 21-day cycle) is frequently used as a single agent.
Gemcitabine appears to cross the blood brain barrier and may be a good option in patients with a history of central nervous system metastases . Alopecia and gastrointestinal toxicity are mild, and it is not associated with significant neuropathy. Gemcitabine is well tolerated and active in metastatic breast cancer, though when gemcitabine was directly compared with weekly epirubicin as first-line chemotherapy in women not previously exposed to an anthracycline, it resulted in a significantly shorter time to progressive disease and a lower OS [39-41]. Thrombocytopenia can be a dose-limiting toxicity, especially in heavily pretreated patients.
Ixabepilone is an epothilone, a class of non-taxane tubulin polymerizing agents that have activity in taxane-resistant patients. As single agent treatment, ixabepilone (40 mg/m2 every 21 days) resulted in an ORR of 19% with a median duration of response of 5.7 months in a clinical trial . Median OS was 8.6 months. Grade 3 and 4 peripheral sensory neuropathy occurred in 14% of patients.
Some data suggest that ixabepilone may have less activity when compared with the taxanes, although it may be better tolerated. In the CALGB 40502 trial discussed earlier, weekly ixabepilone resulted in a shorter median PFS compared with taxanes (7.6 months versus 10 months with paclitaxel and nab-paclitaxel) and OS (21 versus 26 and 27 months, respectively) but resulted in a lower incidence of hematologic toxicity (12 versus 21 and 51%) . Of note, the incidence of serious (grade 3/4) sensory neuropathy was equivalent between ixabepilone and nab-paclitaxel (25% in both arms). Further results of CALGB 40502 are discussed above.
In the presence of mild to moderate hepatic impairment, ixabepilone doses should be adjusted. Its usefulness in later line therapy is often limited by its toxicities of neuropathy, anemia, and fatigue. However, epothilones may cross the blood brain barrier , suggesting it may be an option for patients with central nervous system disease.
Despite FDA approval, it is not available in Europe because the European Medicines Agency (EMA) Committee for Medicinal Products for Human Use (CHMP) concluded that the benefit was marginal at best and the risk of peripheral neuropathy to be significant . Alternate dosing schedules are also under active investigation [46,47].
Oral etoposide (50 mg/m2 daily for 21 days every 28 days) is a reasonable choice, especially for patients with slow-growing disease who desire an oral agent. Etoposide has shown an ORR of 30% in pretreated patients, but may produce hematologic and gastrointestinal toxicity [48-50].
Carboplatin and cisplatin are rarely used as single agents in metastatic breast cancer. Available data suggest the response rate to cisplatin is higher among chemotherapy naive patients rather than in women who were previously treated (ORR 42 to 54% versus less than 10%, respectively) . However, there is renewed interest in using these agents as part of a combination regimen, particularly to treat tumors where DNA damage repair pathways are impaired (such as in women harboring germline BRCA1 mutations and patients with triple-negative breast cancer).