The goal of adjuvant systemic therapy for early-stage breast cancer is to eliminate micrometastatic disease, thus preventing future recurrences. Adjuvant systemic therapy, chemotherapy, targeted therapy, and/or endocrine therapy, has led to a substantial decline in breast cancer mortality in women with operable breast cancer. Although chemotherapy can be effective in preventing distant failure, not all patients benefit. Many patients would remain disease free without therapy, and others would relapse despite treatment. The decision to use adjuvant chemotherapy involves a careful assessment of its benefits and potential risks. New insights into the biology of breast cancer have provided improved tools to predict more accurately the benefits of chemotherapy in a given patient, thus allowing chemotherapy to be used in patients who most likely will benefit and sparing those who are less likely to benefit. This chapter focuses on adjuvant systemic chemotherapy; the benefit of targeted therapy and endocrine therapy are detailed elsewhere in this text.

The Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) has provided key insights about the benefits of adjuvant chemotherapy. The EBCTCG was formed to perform meta-analyses of randomized trials in the adjuvant setting and has been updated approximately every 5 years. The 2000 Overview Analysis (published in 2005) reaffirms the role of adjuvant chemotherapy in the treatment of early-stage breast cancer.1 This analysis involved 28,764 women treated in 60 trials that randomized patients to combination chemotherapy versus no chemotherapy and demonstrated a significant improvement in both the rate of recurrence (relative risk [RR]: 0.77) and breast cancer mortality (RR: 0.83). The relative and absolute risk reductions for both recurrence and death from adjuvant chemotherapy were similar in women with node-negative or node-positive disease, although the absolute benefits were greater in women with node-positive disease. The reduction in risk of recurrence with chemotherapy is seen within the first 5 years after randomization and is maintained at the 10- and 15-year time points.

The absolute benefit of adjuvant chemotherapy in a given patient requires knowledge of that patient’s baseline risk of recurrence as well as the expected relative risk reduction from chemotherapy. A patient’s baseline recurrence risk is a function of both the anatomic extent of disease (ie, tumor size and lymph node status) and the molecular characteristics of the tumor (ie, tumor grade and hormone receptor [HR] and human epidermal growth receptor 2 [HER-2] status). The RR reduction from chemotherapy (ie, a tumor’s sensitivity to chemotherapy) appears to be largely based on the molecular characteristics of the tumor, independent of anatomic features, as discussed later.

Considering a patient’s baseline risk and the relative benefit from chemotherapy identifies groups of patients with markedly different benefits from chemotherapy. Patients who have high baseline risk and chemo-sensitive cancers receive the greatest absolute benefit from chemotherapy, whereas those with low baseline risk and relatively chemo-insensitive cancers receive the least benefit from chemotherapy. Those patients with high baseline risk and relatively chemo-insensitive cancers and those with low baseline risk and chemo-sensitive cancers are likely to derive a modest benefit.

It is now clear that the different subtypes of breast cancer, defined by HR (estrogen and progesterone receptor [ER and PR]) expression and HER-2 status, have differing sensitivity to adjuvant chemotherapy. A recent retrospective analysis suggested that women with HR-positive breast cancer derive significantly less benefit from adjuvant chemotherapy compared with those with HR-negative breast cancer.2 This analysis included 3 randomized trials conducted by the Cancer and Leukemia Group B (CALGB) involving 6644 women with node-positive breast cancer. Results from this study demonstrated 23% more ER-negative patients survived to 5 years without a recurrence if they received more intensive adjuvant chemotherapy compared with only 7% of patients with ER-positive tumors. The results of the Overview Analysis also support this finding, suggesting that absolute survival benefit from adjuvant chemotherapy is greater in ER-negative disease compared with ER-positive disease.1

A subsequent analysis of CALGB 9344 that included HER-2 status provided even more precise delineation of chemosensitivity by tumor subtype. This study evaluated the addition of paclitaxel to 4 cycles of doxorubicin and cyclophosphamide (AC) in patients with node-positive disease.3 HER-2 positivity was associated with benefit from the addition of paclitaxel after adjuvant treatment with AC independent of ER status.4 In contrast, cancers that were HER-2 negative and ER-positive derived little benefit from the addition of paclitaxel.

It is worth noting that data from the overview meta-analysis initially suggested patient age predicted benefit from chemotherapy. For women less than 50 years of age, adjuvant chemotherapy reduced the risk of recurrence and death from breast cancer by 37% and 30%, respectively.1 For women 50 to 69 years of age, the risk of recurrence and death from breast cancer was decreased by 19% and 12%, respectively. However, in an updated analysis, focusing only on patients with HR-negative cancers, benefits from chemotherapy were similar in older and younger women.5 This finding suggests that the earlier analysis may have been influenced by the fact that HR-negative breast cancer is much more common in younger women then in older women and that if controlled for tumor biology, the impact of age on chemotherapy benefit is uncertain.

Although receptor status can be helpful in making decisions about chemotherapy, as described earlier, baseline risk factors such tumor size, lymph node status, and tumor grade must also be taken into consideration. A Web-based program, Adjuvant! Online, is available to help in estimating the risk of disease recurrence and death for patients with and without adjuvant chemotherapy.6 This tool uses information derived from SEER databases, meta-analyses, and individual studies. Population-based validation studies have confirmed the value of this program.7

In addition to computerized modeling, genomic assays of tumor samples may also help in the selection of the optimal adjuvant treatment approach. The Oncotype DX assay is a 21-gene panel performed on paraffin-embedded tumor tissue using reverse transcriptase-polymerase chain reaction analysis. This assay uses the expression pattern of 16 breast cancer–associated genes to derive a recurrence score (RS) based on their expression profile relative to 5 reference genes. RS were initially obtained in 668 tumor blocks from patients with node-negative breast cancer who were treated with tamoxifen on NSABP B-14.8,9 RS was divided into low- (RS <18), intermediate-(18-30), or high-risk (≤31), and Kaplan-Meier estimates of 10-year recurrence rates were 7%, 14%, and 31%, respectively. In addition, they were able to demonstrate a significant relationship between RS and overall survival. Further analysis was done using 651 patients enrolled in NSABP B-20.10 This trial randomized patients with node-negative ER-positive breast cancer to tamoxifen alone or tamoxifen with chemotherapy. Distant recurrence-free survival (DRFS) rates were not significantly different in patients with low RS treated with tamoxifen with or without chemotherapy; however, DRFS was significantly worse in patients with high RS who received tamoxifen alone without chemotherapy. This suggests that chemotherapy is not beneficial for patients with node-negative ER-positive breast cancer receiving tamoxifen with a low RS, but it does provides significant benefit to those with a high RS. The benefits of adjuvant chemotherapy in individuals with intermediate RS are unclear. To better define the benefits of adjuvant treatment for those with intermediate scores, the TAILORx study is randomizing these patients to hormonal therapy alone or chemotherapy followed by hormonal therapy.11 Of note is the fact that the RS criteria for an intermediate score were changed from 11 to 25 for the trial. The study plans to enroll at least 10,000 women with HR-positive, HER-2-negative, node-negative breast cancer. However the trial results will likely not be available until 2013. Although data from the NSABP B-14 and B-20 indicate that the Oncotype DX test can be used to help determine if adjuvant chemotherapy may be beneficial in women with ER-positive node-negative tumors, there are now recent studies suggesting that this test may also be helpful in node-positive patients.12,13

Other genomic tests are available in addition to Oncotype DX, however, these tests require fresh-frozen tissue, making their applicability more limited.14 The Mammoprint assay is a 70-gene assay that is offered as a prognostic test for women less than 61 years of age with either ER-positive or ER-negative breast cancer.15 The test results are dichotomous and report either a high or low risk of recurrence. It has not yet been studied if the assay can also predict sensitivity to adjuvant therapies. The Microarray in Node-Negative Disease May Avoid Chemotherapy (MINDACT) trial is a prospective trial of primarily lymph node-negative breast cancer that opened in August 2007.16 All patients are assessed by prognostic factors included in Adjuvant! Online and by the 70-gene Mammoprint assay. If both assays predict a high-risk status, the patient receives adjuvant cytotoxic chemotherapy and also hormonal therapy if ER positive. If both assays indicate a low risk, no chemotherapy is given, and ER-positive patients are given hormonal therapy only. When there is discordance between the prediction of risk as assessed by Adjuvant! Online and the Mammoprint assay, patients are randomized to receive treatment based on either the Mammoprint or Adjuvant! Online results.

In addition to these tools to help clinicians predict risk of recurrence and possible benefit from adjuvant therapies, practice guidelines are put together by 2 major groups: the National Comprehensive Cancer Network (NCCN) and the St. Gallen Consensus.17 The NCCN guidelines suggest consideration of chemotherapy for women whose breast cancers are 1 cm or more or node positive, regardless of receptor status. These guidelines recommend taking into consideration tumor grade and the presence of lymphovascular invasion for patients with tumors between 0.6 to 1.0 cm to determine if adjuvant chemotherapy should be administered. Trastuzumab is recommended, in addition to adjuvant chemotherapy for women with HER-2-positive tumors 1 cm or more or node positive.

Chemotherapy may be administered prior to surgery or after surgical resection. Although neoadjuvant chemotherapy was initially used for locally advanced breast cancer, it has become more common to use in patients with operable breast cancer. This approach allows more individuals to undergo breast-conserving therapy (BCT) and also allows the observation of responses to systemic treatment within the breast. The foundation for the use of preoperative chemotherapy in operable breast cancer came from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-18 trial, which randomized women to 4 cycles of AC either before or after surgery and found no statistically significant differences in disease-free survival (DFS) and overall survival (OS) between the 2 groups.18,19 Preoperative therapy reduced tumor size in about 80% of women, and significantly more were able to undergo BCT (68% vs 60%; p = 0.001). These findings have been confirmed in other preoperative studies, including NSABP B-27.19,20 Although it is true that preoperative treatment allows tumor response in the breast to be monitored during treatment, it should be noted that there are no clear data to indicate how that information can be used to influence further treatment in that individual.

If chemotherapy is administered postoperatively, therapy is generally administered 4 to 6 weeks after surgery. A retrospective analysis found there was no statistically significant benefit to DFS or OS to starting chemotherapy less than 3 weeks compared with more than 3 weeks after surgery.21 There is, however, data to suggest that a delay to chemotherapy of more than 12 weeks reduces both DFS and OS.22

Many women who receive adjuvant chemotherapy also require radiation therapy. Concurrent administration of chemotherapy and radiation therapy is not recommended due to concerns for increased toxicity. There is some theoretical concern that delaying radiation until after chemotherapy may lead to a higher incidence of local recurrence, particularly with the use of more prolonged chemotherapy regimens. A prospective randomized trial was conducted that randomized women to postoperative radiation either before or after chemotherapy, and it found no significant difference in rates of locoregional or distant recurrence, or death.23,24 Given these data, radiation therapy is generally not initiated until completion of adjuvant chemotherapy.

Just as radiation therapy should be delayed until completion of chemotherapy, hormonal therapy also should not be initiated until after completion of chemotherapy for patients with HR-positive breast cancer. The Southwest Oncology Group (SWOG) 8814 study of postmenopausal women with node-positive, ER-positive breast cancer randomized 1477 women to 1 of 3 arms: tamoxifen alone for 5 years; 6 cycles of cyclophosphamide, doxorubicin, and 5-fluorouracil (CAF) with concurrent tamoxifen; or 6 cycles of CAF followed by tamoxifen. This study demonstrated that the 10-year DFS (60% vs 53%) and OS (68 vs 62%) were superior in patients given sequential chemotherapy and tamoxifen compared with those given concurrent therapy. As a result of these data, sequential rather than concurrent therapy has been adopted as the standard of care when both chemotherapy and hormonal therapy are administered.25

The benefits of combination chemotherapy for the adjuvant treatment of early-stage breast cancer were initially established in the 1970s by Bonadonna and colleagues.26 They randomized lymph node-positive patients after mastectomy to 12 monthly cycles of cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) or to observation and found that CMF results in a significant improvement in DFS and OS. A subsequent study evaluated the possibility of reducing the duration of CMF and found that 6 cycles of CMF yielded identical results to those obtained with 12 cycles.27 A long-term analysis with a median follow-up of 25 years confirmed that longer duration of treatment did not improve treatment outcome, with an estimated DFS of 39% after 12 cycles and 38% after 6 cycles.28

Over the next several years, multiple randomized trials evaluated the benefit of anthracyclines in adjuvant therapy. The National Surgical Adjuvant Breast and Bowel Project (NSABP) B-15 and B-23 studies found that 4 cycles of AC was equivalent to 6 cycles of CMF with regard to DFS and OS.29,30 The 2000 Overview Analysis compared CMF-based and anthracycline-based polychemotherapy and found that among the 14,000 women they examined, there was a small but significant improvement in DFS and OS favoring anthracycline-containing regimens.1 Although 4 cycles of AC has not directly been shown to be superior to CMF, AC chemotherapy has become a commonly used regimen, likely in part due to the shorter duration of this regimen.

Multiple randomized trials have examined the benefit of adding a taxane (either paclitaxel or docetaxel) to anthracycline-based adjuvant chemotherapy (Table 87-1). Two large randomized trials have examined the use of paclitaxel in addition to anthracycline-based chemotherapy, and both of these studies suggest a survival benefit for the addition of paclitaxel. CALGB 9344 found that the addition of paclitaxel to AC was associated with an improvement in 5-year DFS (70% vs 65%) and OS (80% vs 77%).3 An unplanned subgroup analysis suggested that the benefit may be restricted to ER-negative breast cancer. NSABP B-28 was a similar study that randomized 3060 node-positive patients to 4 cycles of AC with or without sequential paclitaxel, and demonstrated an improvement in 5-year DFS (76% vs 72%) but no difference in OS.31 As opposed to CALGB 9344, this study did not demonstrate a difference in benefit based on ER status.