Abstract
The continued improvement in the rates of breast cancer survival over the last decade is owing to a combination of earlier diagnosis through breast screening programs and better public awareness, as well as the widespread introduction of effective systemic therapy following resection surgery as investigated in this recent paper by Albain et al. [1]. The aim of this systemic therapy is to treat any undetected local and distant disease that, if left untreated, may subsequently develop into a life-threatening recurrence. In 1976, Bonadonna et al. demonstrated a significant improvement in disease-free survival with the use of chemotherapy consisting of the alkylating agent cyclophosphamide in combination with two antimetabolites – methotrexate and fluorouracil (CMF) [2]. As a result, CMF treatment became the standard adjuvant chemotherapy used in higher risk women with early-stage breast cancer [3]. Subsequently, anthracyclines were incorporated into adjuvant chemotherapy regimens and a later meta-analysis showed that anthracycline-based regimens, such as cyclophosphamide, doxorubicin and fluorouracil (CAF), confer a moderate but significant advantage over CMF [4].
Tamoxifen is a selective estrogen receptor (ER) modulator that binds to the ER and blocks the proliferative actions of estrogen on the mammary epithelium [5]. Numerous clinical trials and meta-analyses have demonstrated that in patients with ER-positive breast cancers, tamoxifen significantly reduces recurrence rates and improves survival. For this reason, tamoxifen became the mainstay of treatment for ER-positive breast cancer for many years. The largest meta-analysis only reported a marginal benefit of the addition of chemotherapy in older women compared with tamoxifen alone. It also highlighted that no large trials of concurrent versus sequential chemoendocrine treatment had been performed. Any advantages conferred by combining hormonal and cytotoxic therapies, such as reducing breast cancer recurrence, must be balanced against the increased morbidity and mortality associated with these treatments; for example, tamoxifen has estrogenic effects on the uterine epithelium in postmenopausal women and is associated with increased rates of endometrial carcinoma [6]. Chemotherapy regimens that include anthracyclines are associated with myelodysplasia, neutropenic sepsis, cardiac damage and increased rates of leukemia [7]. It has been postulated that the concurrent use of tamoxifen may attenuate the efficacy of anthracyclines; tamoxifen reduces the proliferation rate of hormone-sensitive breast cancer cells and, in doing so, may make them less sensitive to the cytotoxic activity of chemotherapy [8]. This latter concern is supported by experiments utilizing in vitro cell cultures [9].
There was therefore a need to undertake a clinical trial in order to answer two important questions: what is the magnitude of benefit in combining adjuvant chemotherapy and endocrine therapy in postmenopausal women with breast cancer; and, furthermore, if the combination is superior, should tamoxifen be given concurrently with chemotherapy or sequentially?
Results
In their paper, Albain and colleagues report the 10-year results of a study performed by the University of Michigan-based South West Oncology Group (SWOG), which recruited 1558 women throughout the USA and Canada between 1989 and 1995. Inclusion criteria required participants to be postmenopausal with pathological stage T1–3, N1–2 infiltrating adenocarcinoma of the breast. Cancers had to be either ER- or progesterone receptorpositive, or both, and patients had to have no radiological or biochemical evidence of residual disease. Patients were randomly assigned to one of three groups at a 2:3:3 ratio; each group received either tamoxifen treatment alone, CAF with concurrent tamoxifen (CAFT) or CAF followed by tamoxifen (CAF-T). CAF was given four-times weekly for six cycles (dose of doxorubicin: 30 mg/m2) and the dose and duration of tamoxifen was 20 mg orally for 5 years. Participants were reviewed at least twice each year for 8 years, after which time they were then reviewed annually. The primary end point was disease-free survival from the point of registration with the trial. Secondary outcomes included overall survival (time from registration until death from any cause) and adverse event rates.
Of the 1558 randomized women, 1477 were eligible for analysis and 81 were found to be ineligible owing to incorrect or incomplete staging. A total of 361 women were assigned to tamoxifen alone, 550 were assigned to CAFT and 566 were asigned to CAF-T. Age, ethnicity, axillary node involvement and surgical intervention were well balanced throughout all three groups. A total of 15% of the patients who were prescribed CAF did not complete the full course of CAF (owing to toxicity, disease progression or death), and 5% of the total study population stopped tamoxifen early owing to toxicity.
The first analysis compared tamoxifen alone against both groups that received a combination of CAF and tamoxifen with the end points of disease-free survival and overall survival. A total of 50% of women in the tamoxifen-alone group experienced a disease event (179 events in 361 participants) compared with 41% in the tamoxifen plus CAF group (458 events in 1116 participants). This was found to be significantly different using a stratified log-rank test (p = 0.002). The 10-year disease-free survival estimates in the two CAF plus tamoxifen groups were 57% (95% CI: 53–60) compared with 48% (95% CI: 42–53) for the tamoxifen-alone group. The hazard ratio (HR) for disease-free survival in the CAF plus tamoxifen groups compared with the tamoxifen-alone group was 0.76 (range: 0.64–0.91). Overall survival was also statistically significantly greater for combined tamoxifen and chemotherapy (stratified log-rank test: p = 0.043). The 10-year overall survival estimates were 65% (95% CI: 62–68) for the CAF plus tamoxifen groups and 60% (95% CI: 54–65) for the tamoxifen-alone group. Compared with the tamoxifen-alone group, the HR for overall survival in the CAF plus tamoxifen groups was 0.83 (0.68–1.01; p = 0.057).
The second analysis compared CAF-T with CAFT for the end points disease-free survival and overall survival. Disease-free survival was marginally improved for the CAF-T group compared with the CAFT group (stratified log-rank test: p = 0.055). The 10-year disease-free estimates were 60% for CAF-T and 53% for CAFT. The HR for 10-year disease-free survival for CAF-T compared with CAFT was 0.84 (range: 0.70–1.01; p = 0.061). The difference in overall survival was not significant (stratified log-rank test: p = 0.27). Secondary three-way comparisons between the groups demonstrated an absolute 10-year benefit in disease-free survival of CAF-T and CAFT over tamoxifen alone of 12 and 5%, respectively.
Morbidity rates in the CAF plus tamoxifen groups were higher than in the tamoxifen-alone group. Neutropenia, hyperemesis and stomatitis were observed in 44, 23 and 27% of participants, respectively. Thromboembolic events, congestive cardiac failure, myelo-dysplastic syndrome and acute myeloid leukemia were also more common in women who were receiving chemotherapy.
Significance
This study has demonstrated that the addition of CAF chemotherapy to tamoxifen treatment significantly improves the 10-year disease-free survival rate for postmenopausal women with hormone receptor- and lymph node-positive breast cancer. There also appears to be fewer recurrence events if the tamoxifen is given sequentially after CAF rather than concurrently. Despite the increased morbidity and adverse events, which could increase non-breast cancer deaths, chemotherapy conferred a significant 10-year overall survival benefit.
One issue is that breast cancer treatments change every few years as new drugs become available, so by the time long-term results from studies become available, their results have less of an impact. Owing to the introduction of aromatase inhibitors, tamoxifen use in post-menopausal women is declining. The Breast International Group (BIG) 1–98 and Arimidex, Tamoxifen, Alone or in Combination (ATAC) trials have shown superior disease-free survival in postmenopausal women with ER-positive disease with aromatase inhibitors compared with tamoxifen [10,11]. Aromatase inhibitors have become the mainstay of endocrine treatment in postmenopausal women, particularly among those who are node positive. Aromatase inhibitors are even more effective than tamoxifen in reducing proliferation [12], and although one cannot extrapolate the findings from the current study to other endocrine agents with certainty, the theoretical argument for sequencing chemotherapy followed by endocrine treatment is compelling and is now the current recommendation. Moreover, the second-generation anthracycline, epirubicin, has recently been introduced, which causes less cardiac toxicity than doxorubicin; it can be used at higher doses and may result in lower rates of breast cancer recurrence than doxorubicin [101].
Future perspective
Chemotherapy is a blunt tool that reduces mortality in a few individuals and causes significant morbidity in many. Future development must focus on improving the identification of patients at a low risk for recurrence, who have little to gain from adjuvant systemic therapy. Within the hormone receptor-positive, lymph node-positive population, there will also be many women who will be cured by surgery and endocrine therapy alone. Although the 70-gene signature and 21-gene recurrence scores are available, they are essentially prognostic signatures and do not predict which women will benefit from treatment. Using the 21-gene recurrence score, patients with low recurrence scores received no apparent additional benefit from chemotherapy [13], and patients with high recurrence scores had improved disease-free survival if they received additional chemotherapy compared with participants who received tamoxifen alone. High-risk women have greater absolute benefits from chemotherapy, but the majority of relapses are not prevented and the effective identification of patients who will and will not benefit from chemotherapy is therefore urgently needed. This would reduce the overtreatment and unnecessary exposure of patients to toxic therapy that is unlikely to alter the course of their disease.
Further stratification of breast cancer subtypes and genome analysis bring the prospect of individually tailored therapy ever closer. Good patient management will depend on more than simply elucidating the likely disease course for the patient. It is of importance for oncologists to continue to provide honest and balanced advice regarding the benefits and harms of any particular treatment. Through understanding the impact on quality of life as well as the patient's longevity, the patient and oncologist can make an informed decision on the best course of treatment for that individual.
Executive summary
For postmenopausal women with endocrine-responsive, node-positive breast cancer, the addition of cyclophosphamide, doxorubicin and fluorouracil (CAF) chemotherapy to tamoxifen treatment increases the 10-year disease-free survival rate from 48 to 57%.
By treating patients sequentially with tamoxifen after CAF chemotherapy, disease-free survival improved further to 60%; however, this was of marginal significance statistically (p = 0.055).
The addition of CAF chemotherapy increased the risk of morbidity and mortality from recognized side effects.
By developing new tools to further stratify the patient's risk of future recurrence, we may reduce overtreatment in patient groups who are unlikely to benefit from further intervention.
Footnotes
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.