If an oncologist practicing in 1964 at the time of ASCO’s founding was placed in a time capsule and transported to the 2014 ASCO Annual Meeting, she could be forgiven for thinking she had arrived in another universe. During this period, the landscape of breast cancer management across the spectrum of risk assessment and prevention, surgery, radiation, and adjuvant systemic therapy has changed so dramatically as to be virtually unrecognizable. In the past 10 years, there has been a rapid acceleration of our understanding of breast cancer biology, which has fueled new approaches to treatment. Although we have not cured all breast cancer, the uptake of mammographic screening in the industrialized world, coupled with improvements in therapy, has resulted in decreasing mortality rates at a time when the morbidity of treatment is also decreasing.1 A discussion of the advances in any single area of breast cancer management could fill an entire paper. This article will focus on a high-level overview of major advances across the spectrum of treatments for patients with nonmetastatic breast cancer.
Risk Assessment and Prevention
In the mid-1960s, the understanding of breast cancer risk was rudimentary. Although it was recognized that a mother or sister with breast cancer significantly increased the risk of developing the disease, the first large, population-based study examining the level of risk associated with family history of the disease was not published until 1981. The identifications of mutations of the BRCA12 and BRCA23 genes in families with a very high incidence of breast and ovarian cancer, followed by the development of a commercially available test for these mutations, ushered in a new era of risk assessment and management for women with the highest level of risk. Progress in our understanding of other factors contributing to breast cancer risk led to the development of a number of models (e.g., Gail and Tyrer-Cuzick) for quantitating risk in women without a family history suggestive of genetic predisposition. The net result of our improved understanding of risk was an end to the era when women were advised to undergo bilateral prophylactic mastectomy (BPM) because their breasts were lumpy or because they had undergone multiple biopsies showing nonproliferative changes—factors not associated with any substantial elevation in breast cancer risk. Surgery remains an appropriate option for risk reduction for women with the highest levels of risk. BPM reduces the risk of breast cancer development in women with BRCA1 and BRCA2 mutations by greater than 90%,4 whereas bilateral salpingo-oophorectomy reduces the risk of breast cancer by approximately 50% and the risk of ovarian cancer by approximately 80%.5 MRI has also been shown to be a more effective screening tool than mammography or ultrasound in this high-risk group.
An additional option for risk reduction became available when we entered the era of breast cancer chemoprevention in 1998, with the publication of the National Surgical Adjuvant Breast and Bowel Project (NSABP) P1 trial. This trial demonstrated that 5 years of tamoxifen reduces the incidence of estrogen receptor (ER)-positive breast cancer by 50% in women at increased risk of breast cancer development.6 Subsequent studies have shown that raloxifene, a failed breast cancer therapy now used for osteoporosis prevention, and aromatase inhibitors are also effective in reducing the incidence of ER-positive breast cancer. At present, there are no clinically proven options for risk reduction for ER-negative cancers other than prophylactic surgery.
In the 1960s, radical mastectomy was the standard surgical approach to breast cancer. This changed dramatically with the landmark NSABP B04 trial, which demonstrated no difference in survival among women with clinically node-negative disease who were undergoing radical mastectomy, total mastectomy with nodal irradiation, or total mastectomy with delayed axillary lymph node dissection (ALND) if nodal metastases developed.7 This study validated the “systemic disease hypothesis” of breast cancer, which suggested that variations in locoregional therapy would have little effect on survival, opening the door to the standard use of the modified radical mastectomy and providing support for trials of breast-conserving therapy (BCT). Six randomized trials of BCT with whole-breast irradiation versus mastectomy have demonstrated no survival differences, even with long-term follow-up.8 Over time, rates of local recurrence after BCT have progressively declined and no longer differ from those seen after mastectomy: approximately 2%-3% at 10 years for ER-positive tumors, and 6%-8% at 10 years for ER-negative tumors.9
Surgery of the axilla has also undergone dramatic changes. ALND as a staging procedure was replaced by sentinel lymph node biopsy (SLNB), a much less morbid technique for identifying women with axillary nodal involvement. Although SLNB was readily accepted as an alternative to ALND for node-negative disease management, the American College of Surgeons Oncology Group (ACOSOG) Z0011 study, which examined the need for ALND in patients with one or two positive axillary nodes undergoing BCT, generated considerable controversy. This study was built on the findings of NSABP B04, which showed that only one-half of the patients with axillary nodal metastases that were not removed surgically developed axillary-first failure, even in the absence of systemic therapy or radiation therapy. The Z0011 study, performed in women who received both whole-breast radiation therapy (RT) as part of BCT and systemic therapy as standard management of node-positive disease, addressed the question of whether these additional treatments would allow a reduction in the extent of surgery. Compared to SLNB alone, ALND did not improve survival or reduce locoregional recurrence but was associated with significantly increased morbidity.10 A prospective study has found that approximately 85% of patients with clinically node-negative disease who undergo BCT and who are found to have sentinel node metastases can be spared ALND.11 The replacement of radical mastectomy, an inpatient procedure that is disfiguring, causes permanent sensory changes to the chest wall, and has a significant risk of lymphedema development with lumpectomy and SLNB, a brief outpatient procedure with far fewer side effects highlights the progress that has been made in local therapy. It is somewhat paradoxical that as outcomes of BCT have improved, an increased rate of mastectomy—particularly contralateral prophylactic mastectomy—is being observed in the United States. Studies indicate that this is due to patient preference rather than physician recommendation,12 and understanding the patient concerns that motivate these choices is an important subject for research to ensure that the surgical morbidity reductions obtained through clinical trials over the past several decades are not lost.
Evolution of Radiation Therapy for Local Disease
In parallel with changes in surgery, there have been many advances in RT. Postmastectomy radiation therapy (PMRT) has been intensively studied. There had long been consensus that PMRT was indicated in patients with involvement of four or more axillary lymph nodes to improve local control, but the demonstration in the 2005 Oxford Overview Analysis that the use of RT after mastectomy or BCT improved survival outcomes led to a fundamental change in our views regarding the role of local therapy in breast cancer management.13 The survival benefit seen with RT indicates that the systemic disease hypothesis does not apply to all patients and that adequate locoregional therapy contributes to survival for both patients with node-positive and node-negative disease. Progress in radiation oncology has resulted in a decrease in the toxicity of treatment with CT-based treatment planning, prone positioning, and breath holding—all employed to decrease dose to normal structures. Randomized trials have also demonstrated the safety of accelerated whole-breast irradiation to decrease the traditional 6-week treatment time for a large number of women undergoing BCT. With 10 years of follow-up, the Canadian and the START A and B trials compared doses of 42.5 Gy in 16 fractions and 39 Gy in 13 fractions, respectively, to conventional whole-breast irradiation with 50 Gy in 25 fractions have demonstrated no significant differences in local control rates and comparable cosmetic outcomes and toxicities.14,15 The role of accelerated partial-breast irradiation to reduce treatment volume to the tumor bed and a limited amount of normal tissue awaits results from randomized trials.
The recognition in the 1950s and 1960s that even the most radical surgery failed to cure many patients with apparently localized breast cancer led to the concepts that clinically occult, micrometastatic disease is present at the time of diagnosis in many women and that survival outcomes would be improved with the postoperative or “adjuvant” use of effective drug therapy. This concept is so central to breast cancer management today that it is difficult to imagine how controversial it once was. A pivotal early study by Bonadonna et al.16 demonstrating that 1 year of cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) improved survival compared to postsurgical observation in patients with node-positive disease spawned multiple other randomized trials that expanded the indications for chemotherapy for management of node-negative disease and identified more effective chemotherapy combinations. The Oxford Overview Analysis has quantified the benefits of chemotherapy in 100,000 women who participated in 123 randomized trials. The 2005 analysis showed that 6 months of anthracycline-based polychemotherapy reduced the annual risk of breast cancer death by 38% for those patients younger than age 50 and by 20% for those aged 50-69, with benefit maintained through 15 years of follow-up. The addition of taxanes led to a further decrease in recurrence (30.2% vs. 34.8%, 2p < 0.00001) and breast cancer mortality (21.1% vs. 23.9%, 2p = 0.0005).17
In the 1970s, the first adjuvant trials of targeted therapy were initiated. The selective ER modulator tamoxifen—and all
endocrine therapies—target the ER protein, which is present in 80%-90% of breast cancers. It was initially thought that tamoxifen would delay relapse but not prolong survival and that benefit would be limited to postmenopausal women. Multiple randomized trials involving more than 80,000 pre- and postmenopausal women aggregated in the Oxford Overview Analysis have now demonstrated that tamoxifen reduces breast cancer mortality in node-positive and node-negative disease (provided that the tumors contain ER) and that 5 years of treatment is more beneficial than 2 years.17 The reduction in the annual odds of death for women younger than age 40 is 39% compared to 37% for those aged 70 and older.18 Subsequent studies have shown that the aromatase inhibitors offer an incremental improvement in disease-free survival (DFS) and an improved toxicity profile for postmenopausal women requiring endocrine therapy.19 The optimal duration of endocrine therapy is the subject of ongoing research, but several studies (e.g., ATLAS, ATTOM, and MA17) now support a duration of at least 10 years. The Oxford Overview Analysis, apart from its specific conclusions, has greatly facilitated improvements in breast cancer management by providing evidence of the public health benefits of modest but consistent treatment effects. By compiling individual patient-level data from multiple trials, it has enhanced our ability to identify small but clinically significant improvements in the outcomes of both local and systemic therapy.
The other dramatic development in adjuvant therapy was the ability to target the human epidermal growth factor 2 (HER2) receptor, initially demonstrated with the recombinant humanized monoclonal antibody trastuzumab. HER2 is overexpressed in approximately 15%-20% of patients with breast cancer. Prior to the availability of HER2-directed treatment, HER2 overexpression was a marker of poor prognosis. After the demonstration by Slamon and coauthors20 that trastuzumab plus chemotherapy prolonged overall survival in patients with HER2-overexpressing metastatic breast cancer, trials in the adjuvant setting that compared chemotherapy plus trastuzumab to chemotherapy alone demonstrated a highly significant prolongation of DFS (hazard ratio, 0.48; absolute decrease, 13%; p < 0.00001) at 4 years as well as an improvement in overall survival of 3.2% (p = 0.0007) for the patients who received trastuzumab plus chemotherapy.21 Efforts to improve on these results have focused on combining trastuzumab with drugs that either block other portions of the HER2 pathway or bind to distinct domains on the receptor. In neoadjuvant studies, the combinations of trastuzumab and lapatinib or trastuzumab and pertuzumab plus chemotherapy both demonstrated higher rates of pathologic complete response (pCR) than single-agent HER2 blockade.22,23 The addition of pertuzumab to trastuzumab and docetaxel increased the rate of pCR from 29% to 45.8% in the Neosphere study.23 Although results from the adjuvant studies confirming that these improvements in pCR translate into improvements in DFS are not yet available, the studies are completed and pertuzumab is the first agent approved by the U.S. Food and Drug Administration for use in the neoadjuvant setting based on the large improvement in pCR seen in the neoadjuvant trials. This approval is an important milestone because the preoperative testing of new agents can involve far smaller groups of patients and provide outcomes within a period of months, rather than the years needed in the postoperative adjuvant setting. This paradigm offers the promise of accelerated time from drug discovery to clinical use.
In addition to being a vehicle for drug development, neoadjuvant therapy has become a standard part of clinical practice. Initially used to render inoperable disease operable, neoadjuvant therapy has been shown in multiple trials to produce the same survival benefit as when it is administered in the postoperative setting.24 Although initial hopes that administration of therapy prior to surgery would improve its efficacy—either through earlier treatment of micrometastases or by allowing “tailoring” of treatment—have not been realized, a pCR to neoadjuvant treatment is recognized as a marker of improved prognosis. Reduction in tumor burden with neoadjuvant therapy also allows women with tumors too large for BCT to undergo the procedure and reduces the need for axillary dissection in women with clinically node-negative disease.
In the past 50 years, major changes in our understanding of the biology of breast cancer have led to progress in therapy. The NSABP B04 trial ended the Halstedian era of breast cancer management and ushered in a time when chemotherapy was believed to be the optimal approach for the majority of patients. The demonstration that breast cancer is not a single disease, but a group of genetically distinct molecular subtypes, contributed to the ending of the era of large, one-size-fits-all adjuvant trials.25 Now we hope to refine therapy based on molecular markers present in individual tumors. The TailoRx, Mindact, and the RxPonder trials have all addressed the question of whether genomic assays can identify patients with ER-positive disease, with and without nodal involvement, who obtain limited benefit from chemotherapy and who are suitable for treatment with endocrine therapy alone or for participation in trials of novel agents. Results of these studies are eagerly awaited, but use of the 21-gene recurrence score (Oncotype DxTM) has already reduced the use of adjuvant chemotherapy in patients with ER-positive, node-negative breast cancer by 13%-34%.26 There is evidence that these genomic scores also predict not only the risk of distant recurrence but also the risk of locoregional recurrence.27 This evidence raises the possibility that in the future, the extent of surgery and radiation may be tailored according to molecular markers. Additionally, although we have treated patients with breast cancer with a combination of surgery, radiation, and systemic therapy for many years, we increasingly recognize that there are synergies among these treatments that can be exploited to decrease the burden of therapy for patients. Radiation, a “local therapy” thought to be important in preventing local recurrence, has a survival benefit,8 whereas systemic therapy significantly decreases locoregional recurrence.28 The optimal integration of these treatment modalities should be a focus of future research.
In summary, it has been an exciting 50 years. Breast cancer care has been transformed from disfiguring surgery and basic radiation to a multimodality effort with preference-based choices for patients, decreased morbidity, and improved survival. The pace of discovery is accelerating, guaranteeing that the next 50 years will continue to see progress in breast cancer prevention, diagnosis, and treatment.