While adjuvant endocrine therapy is the standard of care in patients with tumors that express HRs, the indication for adjuvant chemotherapy in patients with HR-positive disease is undergoing reevaluation. Adjuvant chemotherapy is typically given in sequence with, and prior to, endocrine therapy, as there have been conflicting outcomes with concurrent therapy [22–24]. On average, patients with HR-positive breast cancers derive less benefit from chemotherapy compared to HR-negative tumors [25]. However, there still appears to be a subset of patients with HR-positive tumors that are chemosensitive. The basis for the addition of chemotherapy to adjuvant endocrine therapy is usually guided by the clinician’s estimation of prognosis and assessment of the endocrine- and chemosensitivity of the tumor. Patients for whom systemic chemotherapy should be strongly considered include patients with grade 2 or 3 disease, those with high-risk features based upon the gene signature (such as Oncotype DX and MammaPrint), and patients with a higher disease burden. That said, not every patient with node-positive disease has to be treated with chemotherapy, and multigene signatures are particularly useful in guiding the clinicians in regards to the decision to recommend chemotherapy [26].

The EBCTCG overview reported a benefit in terms of 5-year recurrence-free survival with adjuvant sequential chemoendocrine therapy over endocrine therapy alone with hazard ratios of 0.64 and 0.85 in patients with ER-positive tumors aged <50 and >50 years, respectively [27]. The larger impact of chemotherapy in younger patients may be partially explained by the endocrine effect of chemotherapy on ovarian function [6]. Similar long-term DFS benefits with the addition of chemotherapy to adjuvant endocrine therapy were also noted in the phase III randomized NSABP B-20 and Southwest Oncology Group (SWOG) 8814 trials [26, 28]. The identification of predictors of chemosensitivity in HR-positive tumors has been identified as a key challenge.

There is evidence suggesting an inverse relationship between HR expression and chemotherapy benefit in luminal breast cancers. In a study of postmenopausal women with ER-positive and node-positive cancers from the International Breast Cancer Study Group (IBCSG) Trials VII and 12-93, the addition of adjuvant chemotherapy to endocrine therapy improved the DFS (hazard ratio = 0.81, p = 0.02, median follow-up of 13 years) [29]. Nonparametric subpopulation treatment effect pattern plot (STEPP) analyses demonstrated that this benefit was limited to the patients whose tumors had low to intermediate levels of estrogen expression. The level of ER appears to predict the response to both endocrine and chemotherapies in opposite directions, but these factors may not entirely overlap. It is likely that there are other biological factors that interact with the ER-signaling pathway to determine chemosensitivity. The assumption that chemosensitivity is inversely related to endocrine sensitivity in ER-positive breast cancer is therefore not clearly defined. Tumors that are endocrine sensitive may also be chemosensitive; conversely, endocrine resistant tumors with poor prognostic factors may not always be chemosensitive.

One biomarker that has been examined in chemosensitive ER-positive tumors is Ki67, a marker of cellular proliferation. In an analysis of 1,521 premenopausal and postmenopausal patients with ER-positive tumors from the IBCSG VIII and IX trials, respectively, a high Ki67 index was found to be associated with poorer DFS, but did not predict an OS benefit with the addition of chemotherapy to endocrine therapy [30]. The adjuvant chemotherapy used in these trials was CMF and did not include anthracyclines and taxanes. In contrast, a high Ki67 index was predictive of both outcome and benefit to adjuvant taxane chemotherapy in ER-positive breast cancers in subset analyses of the PACS 01 and Breast Cancer International Research Group (BCIRG) 001 trials [31, 32]. An important caveat of these findings is that these were unplanned subset analyses. At this point, Ki67 should not be used as a basis of recommendation for adjuvant chemotherapy outside of a clinical trial setting, at least in part because the test is not always reliable.

A major advancement in the identification of biomarkers of chemosensitivity in HR-positive breast cancers has been the development of multigene prognostic signatures. These are typically derived from high-throughput analyses of tumor specimens for gene expression patterns and subsequently validated in patient cohorts from clinical trials. These assays have the potential to identify subsets of patients that would benefit from the addition of adjuvant chemotherapy to endocrine therapy. The 21-gene assay, called the Oncotype DX (Genomic Health, Redwood City, CA, USA), provides a recurrence score (RS) that predicts for risk of 10-year distant recurrence. The RS is derived from a complex algorithm calculated on the gene expressions of a preselected list of 16 genes of biological interest, including genes involved in estrogen signaling, cell proliferation, and HER2/neu signaling and 5 reference genes for normalization purposes [28]. The utility of the RS as a predictor of distant recurrence risk at 10 years was initially assessed in the NSABP B-14 trial, in which patients with ER-positive, node-negative breast cancer were randomized to receive either tamoxifen or placebo. The RS was shown to more accurately predict for distant recurrence than conventional clinicopathologic characteristics in the tamoxifen-treated patients [33].

The utility of the RS to accurately predict 10-year distance recurrences was demonstrated in a retrospective analysis of the NSABP B-20 trial, in which patients with ER-positive, node-negative breast cancer were randomized to either tamoxifen or tamoxifen plus chemotherapy. Patients with a low or intermediate RS (defined as <18, and ≥18 and <31, respectively) were found not to benefit from chemotherapy, while those with a high RS (defined as ≥31) derived a significant benefit from chemotherapy [28]. The absolute difference in the 10-year distant recurrence rates with the addition of chemotherapy in these RS groups was an increase of 1.1 % and 1.8 % and a reduction of 28.6 %, respectively. Similar results were obtained in a retrospective analysis of the SWOG 8814 trial, in which postmenopausal patients with ER-positive, node-positive breast cancer were randomized to receive either tamoxifen or tamoxifen plus anthracycline-based chemotherapy [26]. One of the primary strengths of this assay is that RNA may be extracted from archived formalin-fixed, paraffin-embedded tissue, which is the primary mode of preserving tissue in most pathology departments.

Another multigene signature with prognostic utility is the FDA-approved 70-gene MammaPrint signature (Agendia, Amsterdam, Netherlands). Unlike the Oncotype RS assay where genes are preselected, MammaPrint was developed using an unsupervised hierarchical clustering approach whereby the high-risk gene signature predicted a poor clinical outcome in tumors of all subtypes [34]. A retrospective analysis of pooled patient cohorts with ER-positive, node-negative breast cancer demonstrated that the 70-gene score had prognostic value and predicted improved survival outcomes with the addition of chemotherapy to endocrine therapy only in the subgroup of 70-gene high-risk patients [35].

While both the Oncotype and MammaPrint assays were tested retrospectively, the Oncotype RS was evaluated retrospectively in a prospectively assembled clinical trial. For this reason, there is far greater confidence, at this time, that the Oncotype assay can reliably predict which patients will benefit from chemotherapy, and even more importantly, which ones will not. In addition, unlike the Oncotype assay, MammaPrint is performed on fresh-frozen tissue that may limit its feasibility for routine use. Both of these multigene signatures are currently undergoing prospective validation in large ongoing studies (Oncotype RS, TAILORx and RxPONDER trials; MammaPrint, MINDACT trial), which include over 100,000 patients collectively to definitively address their predictive value for chemosensitivity in ER-positive breast cancer [36, 37].

In considering the benefits of adjuvant chemotherapy in patients with HR-positive tumors, it is important to consider common relapse patterns. Patients with HR-positive tumors are at a continued risk of relapse for many years after initial breast cancer diagnosis [38]. More than half of all recurrences among women treated with adjuvant tamoxifen therapy occur between 6 and 15 years after diagnosis, and the greatest benefit with the addition of chemotherapy in DFS was seen primarily within the first 5 years from diagnosis [27]. The limited benefit from chemotherapy in preventing late relapses is also reflected in the DFS patterns of patients with poor prognosis multigene signatures with both the Oncotype RS and MammaPrint assays [28, 39]. Late recurrences and deaths remain a formidable clinical challenge in HR-positive breast cancer, and chemotherapy is unlikely to be the answer to this problem.

The summary recommendations for adjuvant chemotherapy in hormone receptor-positive breast cancer are as follows:

The advent of HER2-directed therapy has revolutionized the management of HER2/neu-positive, early-stage breast cancer. Based on the results of five randomized clinical trials, 12 months of adjuvant trastuzumab is now an integral part of systemic therapy for these patients [40–42]. In all studies, trastuzumab was added to a chemotherapy backbone, and there is currently no data to support the use of adjuvant trastuzumab monotherapy. Evidence-based chemotherapy backbones in this context include AC → T (NSABP B-31, NCCTG N9831), AC → docetaxel (BCIRG 006) and docetaxel + carboplatin (BCIRG 006). Given the increased cardiotoxicity risk upon administering trastuzumab concurrently with an anthracycline in the metastatic setting [43], trastuzumab is omitted during the period of anthracycline chemotherapy.

There remains controversy about the treatment of small HER2/neu-positive cancers. There are limited data on outcomes for patients with small, stage I HER2/neu-positive breast cancers because the seminal adjuvant trastuzumab trials excluded patients with these tumors. Current guidelines from St. Gallen and the European Society for Medical Oncology (ESMO) do not recommend adjuvant trastuzumab and chemotherapy for node-negative HER2/neu-positive tumors that are <1 cm [44]. In contrast, the National Comprehensive Cancer Network (NCCN) treatment guidelines have factored in the indirect evidence obtained from retrospective and subset analyses of trials and recommend consideration be given to the use of trastuzumab-based therapy in T1bN0 tumors, in particular, in the hormone receptor-negative subset [45].

However, there is a wide variation in clinical practice in this subgroup. Recently, interest has developed in using less intensive, and therefore potentially less toxic, partner chemotherapies with adjuvant trastuzumab for low-risk HER2-positive tumors. In a phase II study in women with HER2/neu-positive metastatic breast cancer, weekly paclitaxel and trastuzumab resulted in a 67–81 % response rate, and a 6 % incidence of grade 3 or 4 neutropenia [46]. The Dana-Farber Cancer Institute led a multicenter, phase II, nonrandomized study of weekly paclitaxel plus trastuzumab for 12 weeks, followed by maintenance trastuzumab for a further 9 months in patients with node-negative, HER2/neu-positive tumors that are <3 cm (information available at ClinicalTrials.gov; identifier NCT00542451). This trial has completed accrual of 410 patients, of whom approximately 50 % had tumors <1 cm. If the 3-year DFS is >95 %, the regimen will be deemed worthy of further investigation. The results of this trial were reported at the San Antonio Breast Cancer Conference in 2013. There was a high disease free survival rate of 98.7% at a median follow up of 4 years in the population studied, and there were very few adverse events associated with this regimen [47]. In light of these findings, the combination of paclitaxel and trastuzumab should be considered for patients with stage 1 breast cancer.

The adjuvant therapy of HER2/neu-positive breast cancer will be discussed in detail in a separate chapter. The summary recommendations for systemic adjuvant chemotherapy in HER2/neu-positive breast cancer are as follows:

Inflammatory breast cancer (IBC) represents a unique biological entity characterized by distinct clinical and histopathological features, aggressive behavior, and an exceptionally poor prognosis (median survival with current therapy <4 years) [56]. The current standard of care for management of stage 3B IBC is a multimodality approach consisting of NAC followed by surgery and radiotherapy. Achieving a pCR to NAC is the single most important prognostic factor in IBC [57, 58].

Given the relative rarity of IBC, there have been no specific randomized trials examining the optimal NAC regimen, and moreover patients with IBC have historically been excluded from NAC systemic therapy studies due to their poor prognosis. Single-arm studies and retrospective case series show that anthracycline-based regimens are effective (clinical response rates around 70 %) [57] and that their efficacy is enhanced by the subsequent addition of a taxane as evidenced by increased clinical and pCR rates [59, 60]. As such, regimens included in the “Anthracycline and Taxane” section of Table 23.1 are recommended.

Although outside the scope of this chapter, it is noteworthy that approximately 40 % of IBC are HER2/neu positive, and evidence from randomized phase 3 clinical trials strongly supports the routine addition of trastuzumab to NAC in this setting [61].

Although the incidence of breast cancer rises sharply with age, there is a lack of quality data discussing the optimal choices regarding adjuvant chemotherapy in the elderly. This is particularly true for patients with advanced comorbidities and frailty, who are generally excluded from phase 3 clinical trials. For this reason, groups such as the International Society of Geriatric Oncology (SIOG) and European Society of Breast Cancer Specialists (EUSOMA) have developed consensus guidelines specific to the issues facing elderly patients with breast cancer [62]. Compared to their younger counterparts, elderly patients are more likely to present with larger primary tumors and positive lymph nodes [63], at least in part attributable to greater delays in initial diagnosis. Breast cancers in the elderly are also more frequently HR positive [64].

Prescribing adjuvant chemotherapy to elderly patients require several unique considerations. First, elderly patients often suffer comorbid illnesses that provide competing mortality risks. As such, determining the potential overall survival gains from adjuvant chemotherapy for an individual patient is more challenging. Groups such as the Cancer and Leukemia Group B (CALGB) have developed tools for pre-chemotherapy geriatric assessment to help address this problem [65]. It is also important to note that despite competing risks, >40 % of patients diagnosed with breast cancer after the age of 80 will die from breast cancer [66].

Second, elderly patients may be more susceptible to certain chemotherapy toxicities. Although there is no evidence to support modifying chemotherapy doses because of age, strict monitoring of renal and hepatic function during treatment is essential. Furthermore, routine assessment of left ventricular ejection fraction is recommended for patients scheduled to receive anthracyclines. Thirdly, data suggests that elderly patients are more likely to experience difficulties with medication compliance [63], particularly oral medications such as antiemetics. It is thus critical to ensure that patients with any degree of cognitive impairment clearly understand their drug regimens and are adequately educated and supervised if necessary.

There is little prospectively collected, randomized trial data to suggest a particular adjuvant chemotherapy regimen for elderly patients beyond the general standards of care. In general terms, adjuvant chemotherapy is feasible in patients over 65–70 years of age, but increasing age, reduced functional status, and presence of comorbidities are associated with more frequent dose reductions and/or delays [67]. A landmark CALGB study compared standard chemotherapy regimens (either AC or CMF) to oral capecitabine in patients >65 years of age with early-stage breast cancer [68]. The study was stopped early after an interim analysis suggested that the capecitabine regimen was inferior, resulting in an almost doubled risk of recurrence or death. This study reinforces the efficacy of standard chemotherapy in an elderly population. Nonrandomized data also suggest tolerability of the TC regimen patients >65 years [69], although elderly patients do seem more prone to taxane-induced hematological toxicities [70].