Adjuvant Systemic Chemotherapy in Early Breast Cancer
Chau Dang
Clifford A. Hudis
INTRODUCTION
In women, breast cancer is the most common life-threatening malignancy, but the second leading cause of cancer mortality with a falling mortality rate in the United States over the past few decades (1). This trend has been attributed to both large-scale screening leading to the identification of earlier stage disease with lower risk and to an improvement in systemic treatment strategies that reduce the likelihood of recurrence. Individual large-scale, randomized, controlled clinical trials have generally indicated modest improvements in disease-free survival (DFS) and overall survival (OS) but translated into a large population the public health impact can be substantial. Worldwide overviews conducted by the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) have confirmed the scale of these improvements. Based on these meta-analyses, the combination of chemotherapy and hormonal therapy would be expected to halve the rate of breast cancer mortality in patients with estrogen receptor (ER)-positive disease during the first 15 years after diagnosis (2). Similarly, polychemotherapy significantly improve outcomes in patients with ER-poor breast cancer (3). Recently, the EBCTCG reported long-term outcomes of various polychemotherapy regimens for early breast cancer, including trials that contained the taxanes (4). This chapter will focus on the evolution, and current status of adjuvant chemotherapy in the treatment of early breast cancer.
DEVELOPMENT OF COMBINATION CHEMOTHERAPY
First-Generation Trials with Alkylating Agents and Cyclophosphamide-, Methotrexate-, and Fluorouracil-Based Chemotherapy
The first trial of adjuvant chemotherapy for breast cancer was performed by the National Breast Cancer Bowel and Breast Project (NSABP) in 1958 (5). This study was based on the premise that cancer cells were sometimes dislodged at the time of surgery. Two days of perioperative thiotepa was compared to placebo in over 800 women undergoing radical mastectomy and showed a benefit in premenopausal patients. In an attempt to improve upon these results, the NSABP studied 2 years of postoperative adjuvant therapy with melphalan and again showed a more pronounced benefit in premenopausal, node-positive patients (6).
The observed activity of combination chemotherapy consisting of cyclophosphamide, methotrexate, and fluorouracil or 5-FU (CMF) for metastatic breast cancer led to its incorporation as adjuvant therapy. In 1975 Bonadonna et al. reported that 12 cycles of cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) resulted in a 19% advantage in 5-year DFS compared with no adjuvant chemotherapy (p < .002) and a 14% improvement in OS (p < .04) (7). This combination eventually became known as “oral” or classical CMF because the cyclophosphamide (100 mg/m2) was administered orally for 14 days of the 28-day regimen.
Methotrexate (40 mg/m2) and 5-fluorouracil (600 mg/m2) were both administered intravenously (IV) on days 1 and 8. A follow-up study, comparing 6 months of therapy to 12 months, demonstrated an equivalent outcome (8). Bonadonna et al. reported the 30-year follow-up on randomized studies of adjuvant CMF and showed the enduring benefit of this regimen in the adjuvant setting (9). Similarly, the Cancer and Leukemia Group B (CALGB) conducted several trials of CMF-based therapies and demonstrated that the clinical benefit of chemotherapy was maintained with over 20 years of follow-up (10).
A joint analysis of two studies conducted by the International Breast Cancer Study Group and German Breast Cancer Study Group compared 6 months of CMF to 3 months (11). There was a trend toward a benefit for 6 months of therapy in women 40 years or younger (p = .22) and in those with hormone-receptor-negative breast cancer (p = .37). In the German Adjuvant Breast Cancer Group comparison of 6 months to 3 months of adjuvant treatment with CMF in 789 women there was no difference in outcome (12). The Southwest Oncology Group (SWOG) found that a 2-year duration of CMF with vincristine and prednisone (CMFVP) in patients with axillary-node-positive, hormone-receptornegative breast cancer was no better than 1 year in terms of OS (p = .33) or DFS (p = .24) (13). The Ludwig Trial Group found that one peri-operative cycle of CMF chemotherapy was superior to surgery without any chemotherapy (p = .04) (14). However, a follow-up study was done by the Ludwig group compared one perioperative cycle of CMF to 6 cycles in node-positive breast cancer, and they concluded that CMF for 6 cycles was superior in terms of DFS (p < .0001) and OS (p = .011) (15). Based on the results of these earlier trials of CMF with 6 months of CMF adjuvant therapy being equivalent to longer durations of therapy, this has become the most popular duration of therapy for the three-drug combination. Furthermore, studies utilizing adjuvant CMF in combination with other agents (i.e., prednisone, vincristine) failed to demonstrate a consistent an impact on DFS or OS (10, 16, 17 and 18).
Most of the earlier trials were performed in the nodepositive population. The trial NSABP B-13 evaluated no chemotherapy versus methotrexate/5-flourouracil (MF) in 760 patients with hormone-receptor-negative, node-negative disease and showed a benefit of MF through 16 years of followup in terms of relapse-free survival (RFS) (hazard ratio [HR] = 0.59, p < .001; OS: HR = 0.75, 95% CI = 0.58-0.98, p = .03) (19). In a subsequent trial (NSABP B-19), 1,095 patients with ER-negative, node-negative tumors were assigned to MF versus CMF, and through 13 years of follow-up, an overall benefit was seen in favor of CMF (RFS: HR = 0.59, p < .001; OS: HR = 0.71, p = .01), especially in patients 49 years or younger (19). The NSABP B-20 enrolled patients with hormone-receptorpositive, node-negative disease and with a median follow-up of 8 years, it demonstrated that the addition of chemotherapy (MF or CMF) to tamoxifen was superior to tamoxifen alone in both DFS (84% vs. 77%, p = .001) and OS (92% vs. 88%, p = .018) (20). Based on the results of these nodenegative trials, the 2000 National Institutes of Health (NIH) Consensus Conference recommended that all patients with tumors of greater than 1 cm or positive lymph nodes should be offered adjuvant chemotherapy, regardless nodal status, menopausal status, and hormonal receptor status (21).
Adjuvant Anthracyclines
Doxorubicin
The anthracyclines were introduced in the adjuvant setting, mainly doxorubicin or epirubicin, in the 1980s. In 1981, NSABP B-11 added doxorubicin (A) to melphalan and fluorouracil (PAF) versus melphalan and fluorouracil (PF) in over 700 ER-poor patients (22). The study demonstrated that the addition of doxorubin conferred a DFS (p = .003) and OS (p = .05) benefit. The CALGB 8082 compared a prolonged regimen of cyclophosphamide, methotrexate, fluorouracil, vincristine, and prednisone (CMFVP) with CMFVP followed by anthracycline-based vinblastine, doxorubicin, thiotepa, and fluoxymesterone (VATH) in over 900 patients. At a median follow-up of 11.5 years, the incorporation of VATH led to significant improvements in DFS (p = .004) and OS (p = .043) (23). Bonadonna et al. conducted two trials involving doxorubicin given sequentially with CMF. The first trial compared CMF IV × 12 versus CMF IV × 8 followed by doxorubicin × 4 (CMF → A) in over 550 women with one to three positive nodes. The second study enrolled over 400 patients with four or more involved nodes and evaluated doxorubicin × 4 followed by CMF × 8 (A → CMF) versus alternating 2 courses of CMF with 1 cycle of doxorubicin (CMF/A) (24). There was a benefit in DFS and OS when sequential doxorubicin was given all before CMF in the second trial. This may be due to the importance of maintaining dose density which will be discussed later in this chapter. However, there was no benefit in DFS or OS with the addition of doxorubicin in the first trial, and perhaps this could be due to the inclusion of a lower-risk population in this study (one to three positive nodes).
The US Intergroup conducted a study (INT-0102) comparing CAF × 6 cycles (100 mg/m2 orally days 1 to 14, 30 mg/m2 IV days 1 and 8, 500 mg/m2 IV days 1 and 8) against classical CMF in over 3,900 patients with high-risk (tumor size greater than 2 cm, ER-negative, or high S-phase fraction), node-negative disease, with or without tamoxifen. The study showed that CAF was associated with a marginal improvement in OS (85% vs. 83%, HR = 1.19 for CMF vs. CAF, p = .03) (25). The Grupo Espanol de Investigacion en Cancer de Mama (GEICAM) group enrolled over 980 patients with stages I to III breast cancer and compared FAC (500/50/500 mg/m2) IV versus CMF (600/60/600 mg/m2) IV, all given every 3 weeks × 6, and showed superiority in favor of FAC in DFS (RR 1.2, p = .03) and OS (RR 1.3, p = .05) (26).
The NSABP conducted study B-15 that enrolled over 2,100 patients with node-positive disease to a shorter course of an anthracycline-based treatment with the omission of fluorouracil. In this study patients were randomized to AC (60/600 mg/m2) IV every 3 weeks × 4 versus classical CMF × 6 versus AC × 4 followed by 6 months of rest period followed by CMF IV every 28 days × 3 (750 mg/m2 days 1 and 8, 40 mg/m2 days 1 and 8, 600 mg/m2 days 1 and 8) (AC → rest → CMF) and found no significant DFS or OS in all three arms (27). Next, the NSABP conducted trial B-16 and showed the benefit of the addition of AC therapy with tamoxifen over endocrine therapy alone in node-positive patients (28). Moving forward, the NSABP conducted study B-23 and randomized over 2,000 node-negative patients to classical CMF × 6 versus AC × 4, with or without tamoxifen. Similar to B-15, there was no difference in RFS or OS between the two regimens (18, 29). Unlike other anthracycline-based regimens (CAF or FAC) and epirubicin-based therapies (discussed below), all given either as 6 cycles or over 6 months, AC was not better than classical CMF. There are several reasons to explain the lack of benefit of AC over classical CMF, and they include the short duration of this therapy, the omission of 5-fluorouracil, and the use of intravenous cyclophosphamide. However, the AC regimen was considered to be preferable to the classical CMF regimen due to ease of administration and its convenience, and it was incorporated into several subsequent sequential anthracycline-taxane combinations (Table 44-1).
TABLE 44-1 Trials of Doxorubicin- versus Non anthracycline-Containing Regimens | ||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Epirubicin
The National Cancer Institute of Canada (NCIC) conducted trial MA.5 and enrolled over 700 node-positive patients to cyclophosphamide, epirubicin, and fluorouracil (CEF) (100 mg/m2 orally days 1 to 14, 60 mg/m2 IV days 1 and 8, 600 mg/m2 IV days 1 and 8) versus classical CMF (30). Now at 10 years, results showed a statistically significant improvement in relapse-free survival (RFS) (52% vs. 45%; p = .007) and a trend in OS (62% vs. 58%, p = .085) in patients in the epirubicin-containing arm. Treatment with CEF was accompanied by more toxicities than CMF, including hospitalizations for febrile neutropenia (8.5% vs. 1.1%). Nevertheless, CEF became the standard chemotherapy in Canada for treating high-risk patients.
The French Adjuvant Study Group (FASG) conducted study FASG 05 and randomized 565 patients to FEC 100 (500/100/500 mg/m2) versus FEC 50 (500/50/500 mg/m2), all administered IV every 3 weeks × 6 cycles (31). At a median follow-up of 10 years, results were in favor of FEC 100 in DFS (50.7% vs. 45.3%, p = .036) and OS (54.8% vs. 50.0%, p = .05). Delayed cardiac toxicity after relapse occurred in 4.3% and 4.1% of patients in FEC 50 and FEC 100 arms, respectively. Thus, the survival advantage of FEC 100 was not offset by a higher incidence of cardiac toxicities. At that time FEC 100 then became a standard adjuvant treatment option in Europe.
Poole et al. reported on the combined results of two trials investigating the efficacy of epirubicin in the National Epirubicin Adjuvant Trial (NEAT) and the Scottish trial BR9601 in over 2,300 women (32). In the NEAT study, patients were randomized to E (100 mg/m2) IV × 4 followed by classical CMF × 4 (E → CMF) versus classical CMF × 6. In BR9601, patients were randomized to E (100 m/m2) × 4 followed by modified CMF (750/50/600 mg/m2) IV every 3 weeks × 4 versus modified CMF × 8. Overall, results were in favor of the addition of epirubicin with 5-year RFS rates of 76% versus 69% and 5-year OS of 82% versus 75% (p < .001 for all comparisons). This study confirmed the benefit of an anthracycline-based regimen over CMF alone. Thus, E → CMF regimen, similar to the original Bonadonna regimen, became another standard option as adjuvant therapy in Europe (24) (Table 44-2).
The EBCTCG evaluated 14,000 patients in trials comparing CMF with an anthracycline-based regimen and demonstrated a modest, but statistically significant, benefit favoring the anthracyclines, with absolute benefits of 3.4% for recurrence and 3.3% for mortality at 15 years of follow-up, irrespective of age or hormone receptor status (2). Epirubicin-based regimens are commonly used in Europe and Canada. Epirubicin has almost half of the cardiotoxicity than seen doxorubicin, on a milligram by milligram basis (33). However, the cumulative doses of epirubicin in the adjuvant regimens are higher than those for doxorubicin, and thus, the cardiac toxicity of epirubicin-based regimens are quite similar to that with doxorubicin-containing therapies.
The anthracyclines have been a mainstay of breast cancer treatment for the past 30 years. The use of the anthracyclines has been declining due to various reasons, as specific treatments for different subgroups of women are being defined. However, the data are unclear on which subsets of patients may have the anthracyclines omitted. The literature on response to the anthracyclines, or not, based TOPO II gene aberrations or HER2 overexpression or amplification is still unclear (34). Thus, given the established benefit of the anthracyclines, any decision to eliminate doxorubicin or epirubicin, especially in high-risk patient, should be based on the results of adequately powered, prospective randomized trials.
Despite the use of the anthracyclines, the risk of relapse remains significant, especially in those with poor risk features (35). Inherent or acquired drug resistance could result in disease relapse and variable drug sensitivity among subclones could also contribute to resistance (36). To overcome the limitations of chemotherapy, various strategies have been investigated, which include the manipulations of dose and schedule and the use of non-cross-resistant drugs, such as the taxanes.
Adjuvant Taxanes
Paclitaxel
The CALGB 9344 trial enrolled more than 3,100 node-positive women to receive AC versus AC → paclitaxel (P), and three different doses of doxorubicin were evaluated (60 mg/m2, 75 mg/m2, or 95 mg/m2) (37). There was no benefit with the higher doses of A beyond the standard dose of 60 mg/m2. At a follow-up at 5 years, the hazard reductions from adding paclitaxel to AC were 17% for recurrence (p = .0023) and 18% for death (p = .0064). The DFS was 65% for AC alone versus 70% for AC → P; OS was 77% for AC and 80% for AC→ P. The effects of adding paclitaxel were not significantly different in subsets defined by protocol. However, in an unplanned subset analysis, the hazard ratio (HR) of AC → P versus AC
alone was 0.72 for those with ER-negative tumors and only 0.91 for those with ER-positive tumors. The additional toxicities of P were modest.
alone was 0.72 for those with ER-negative tumors and only 0.91 for those with ER-positive tumors. The additional toxicities of P were modest.
TABLE 44-2 Trials of Epirubicin- versus Non Anthracycline-Containing Regimens | |||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| |||||||||||||||||||||||||
Similarly, the NSABP B-28 also randomized over 3,000 women with node-positive breast cancer to AC × 4 versus AC × 4 followed by P × 4 (38). The 5-year results were reported showing the benefit in the addition of P to AC in all nodepositive breast cancer patients, regardless of ER status, use of tamoxifen, age, number of positive nodes, type of surgery, tumor grade, and histologic type. The 5-year DFS was 72% versus 76% (p = .008) in favor of addition of P. The 5-year OS was 85% in both groups. The lack of OS benefit with P in this study, unlike CALGB 9344, may be attributed to a better prognosis group (70% in NSABP B-28 vs. 46% of patients in CALGB 9344 with one to three positive nodes) and the concurrent use of tamoxifen with chemotherapy in NSABP B-28.
With the results of CALGB 9344 and NSABP B-28, every 3 weeks AC → P became the standard option for the treatment of high-risk breast cancer and became the standard arm in the next trial led by Eastern Cooperative Oncology Group (ECOG) in study E 1199 (39). This study randomized over 5,000 node-positive and high-risk node-negative patients to AC → P every 3 weeks versus AC → weekly P × 12 (80 mg/m2) versus AC → docetaxel (D) (100 mg/m2) every 3 weeks versus AC → weekly D × 12 (35 mg/m2). Overall, when compared to control, AC → weekly P was superior in terms of DFS (81.5% vs. 76.9%, p = .006) and OS (89.7% vs. 86.5%, p = .01), and AC → D was better only in DFS (81.2%, p = .02). The result of this trial confirmed the benefit of the weekly paclitaxel over the every 3 weekly schedule as demonstrated in the metastatic setting (40). In an exploratory analysis, the addition of weekly paclitaxel was beneficial regardless of hormone receptor or HER2 status. As AC → weekly P was better in both endpoints and there were higher rates of febrile neutropenia and infection with AC → D than in other groups, AC → weekly P emerged as the best option from this trial. A direct comparison of weekly versus dosedense (every 2 weekly) adjuvant paclitaxel was performed by SWOG (0221) and results are anticipated in 2013.
Another trial demonstrating the benefit of weekly paclitaxel was conducted by the GEICAM group in study 9906 (41). In this study over 1,200 patients were randomized to standard 6 cycles of FEC (600/90/600 mg/m2) IV every 3 weeks or FEC ×4→P (100 mg/m2) weekly × 8. The 5-year DFS was superior for the FEC → P arm (78.5% vs. 72.1%, p = .006). The FEC → P was associated with a 23% reduction in relapse (p = .022) and 22% decrease in death but this was not statistically significant (p = .110). This translated into a 37% risk reduction in recurrence (p = .0008) in all subsets. There was no significant interaction hormone receptor or HER2 status and paclitaxel treatment. Toxicities were acceptable for both arms. Thus, this was another study strongly confirming the benefit of a taxane in the adjuvant treatment of breast cancer. Recently, the GEICAM group also showed the benefit with the addition of weekly paclitaxel to an anthracycline backbone in a node-negative population (42). In this trial patients were randomized to FAC (500/50/500 mg/m2) every 3 weeks × 6 versus FAC ×4 → weekly P (100 mg/m2) × 8. Overall, the 5-year DFS was in favor with the addition of P (93% vs. 90%, p = .043). Although the result of this study was consistent with most other anthracycline-taxane trials, 58% of the patients had T1 tumors and 73% had hormonereceptor-positive disease. In the era of precision medicine most of these patients should undergo genomic profiling (i.e., ONCOTYPE) to determine if they truly need chemotherapy. This will be discussed later in this chapter.
In terms of administering paclitaxel concurrently with an anthracycline, the European Cooperative Trial in Operable Breast Cancer (ECTO) randomized over 1,350 patients to control of sequential A (75 mg/m2)→ CMF versus concurrent A/P (60/200 mg/m2) → CMF preoperatively or postoperatively (43). The addition of paclitaxel significantly improved RFS over control (HR = 0.73, p = .03), and RFS outcomes were the same whether chemotherapy was given preoperatively or postoperatively. Another trial of concurrent P with A was conducted by the Loesch et al. comparing every 3 week AC → P against concurrent A/P (50/200 mg/m2) → weekly P (80 mg/m2) (44). This study did not show a benefit of concurrent A/P → weekly P. This may be attributed to the higher percentage of patients requiring dose reduction (14% vs. 3%) and delays (14% vs. 5%) in the A/P → weekly P arm. Thus, unlike the result in E1199 showing the benefit of AC → weekly P, the specific regimen (A/P → weekly P) cannot be considered a standard option (Table 44-3).
Docetaxel
Docetaxel is active in the adjuvant treatment of breast cancer. Docetaxel has no pharmacologic interaction with the anthracyclines, thus allowing its coadministration with an anthracycline in an effort to achieve a (theoretical) synergistic response, or to minimize treatment duration. However, toxicity generally means that such a combination will utilize lower doses of these active agents, especially when agents are combined concurrently. One such concurrent combination trial with docetaxel was conducted by the Breast Cancer International Research Group (BCIRG) 001 was a randomized phase III trial comparing FAC with docetaxel plus AC (DAC) in node-positive breast cancer patients (45 and 46). In this study almost 1,500 patients were randomly assigned to treatment with DAC (75/50/500 mg/m2) or FAC (500/50/500 mg/m2), all given every 3 weeks × 6 cycles. Now at a long-term followup of 10 years, the benefit of DAC was maintained for DFS (62% vs. 55%, p = .004) and OS (76% vs. 69%, p = .002) (46). In terms of toxicity, the febrile neutropenia rate was 10-fold
higher with DAC over FAC (25% vs. 2.5%) and thus, granulocyte colony stimulating factor (GCSF) should be used when administering DAC. The GEICAM group evaluated the efficacy of DAC against FAC in over 1,000 patients with nodenegative disease in study 9805 (47). At a median follow-up of 77 months, DFS in favor of DAC (87.8% vs. 81.8%, p = .01), but and OS was not statistically different between the 2 arms. In this study about half of the patients had T1 tumors and two-thirds had hormone-receptor-positive breast cancer. Once again, in the modern era most of these patients should undergo genomic profiling to help determine if they are likely to benefit from chemotherapy.
higher with DAC over FAC (25% vs. 2.5%) and thus, granulocyte colony stimulating factor (GCSF) should be used when administering DAC. The GEICAM group evaluated the efficacy of DAC against FAC in over 1,000 patients with nodenegative disease in study 9805 (47). At a median follow-up of 77 months, DFS in favor of DAC (87.8% vs. 81.8%, p = .01), but and OS was not statistically different between the 2 arms. In this study about half of the patients had T1 tumors and two-thirds had hormone-receptor-positive breast cancer. Once again, in the modern era most of these patients should undergo genomic profiling to help determine if they are likely to benefit from chemotherapy.
TABLE 44-3 Trials of Paclitaxel-Containing Regimens | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related posts:
Breast Cancer Screening
Ductal Carcinoma In Situ and Other Intraductal Lesions: Pathology, Immunohistochemistry, and Molecular Alterations
Postmastectomy Radiation Therapy
Preoperative Endocrine Therapy for Operable Breast Cancer
Management Summary for the Care of Patients with Metastatic Breast Cancer
Nursing Care in Patient Management and Quality of Life
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
