Since the early 1990s, a decline in breast cancer mortality has been observed in both Western Europe and the United States because of the adoption of screening mammography and the widespread use of systemic treatment.1 It has become apparent that improved local control is also causally associated with improved breast cancer survival, a paradigm shift in the understanding of this disease. An absolute reduction in local recurrence at 5 years is associated in a 4:1 ratio with an absolute survival advantage at 15 years.2 This information was not available when recommendations for locoregional treatment were made in 1991.3 It was therefore considered appropriate to amend these recommendations and to consider the impact of improvements in surgery, radiation therapy (RT), and systemic treatments on local control.
Guidelines for the locoregional treatment of primary breast cancer were last published by the US National Institutes of Health in 1991. Since then, new surgical and radiotherapeutic techniques have been developed, clinical trials have provided new evidence, and intriguing long-term effects have emerged from global metadatabases. A revision of these guidelines is therefore necessary and timely. To address this concern, in October 2008, a group of opinion leaders from Austria, Denmark, France, Germany, Italy, the Netherlands, the United Kingdom, and the United States who have worked and published in the field of locoregional treatment met and collectively prepared and approved the described set of recommendations for the use of surgery and radiotherapy in primary breast cancer outside of clinical trials. Cancer 2010. © 2010 American Cancer Society.
Representatives from Europe and the United States (see Table 1) who have worked and published in the field of locoregional treatment were selected by the consensus chairs. The panel included oncologic surgeons, radiation oncologists, pathologists, radiologists, plastic surgeons, medical and gynecologic oncologists, and epidemiologists. Topics for discussion were defined by the panel, and a set of questions was formulated (Table 2). At a meeting in Biedenkopf, Germany in October 2008, 24 presentations provided an overview of current knowledge relevant to the questions being addressed. Presenting panel members were charged with reviewing the most pertinent MEDLINE published reports, as well as abstracts published in the proceedings of the American Society of Clinical Oncology, San Antonio Breast Cancer Symposium, European Cancer Conference, European Society for Therapeutic Radiation and Oncology, European Society for Medical Oncology, American Society for Therapeutic Radiology and Oncology, and European Breast Cancer Conference. The content of the presentations was discussed, the questions were debated, and consensus was achieved. Consensus was based on the results of clinical trials or meta-analyses when available. In the absence of high-quality evidence, consensus was based on expert opinion. The condensed recommendations in this article were approved by all panelists and are summarized in Table 3.
|Harry Bartelink, Netherlands Cancer Institute, Amsterdam, the Netherlands||“Is a Boost to the Tumor Bed Standard for All Patients?”|
|“Can We Agree on Quality Standards for Breast Care?”|
|Thomas A Buchholz, The University of Texas M. D. Anderson Cancer Center, Houston, Tex||“Axillary Irradiation as an Alternative to Dissection”|
|Marco Colleoni, European Institute of Oncology, Milan, Italy||“Timing of Adjuvant Systemic Therapy and Radiotherapy After Breast Conservation and Mastectomy”|
|Sarah Darby, CTSU, Oxford, United Kingdom||“Latest Local Therapy Overview Data”|
|Carsten Denkert, Charite University Hospital, Berlin, Germany||“Histopathologic and Molecular Predictors of the Behavior of DCIS”|
|Jürgen Dunst, University of Lübeck, Lübeck, Germany||“Breast Reconstruction and Radiotherapy: Oncologic Issues”|
|Michael Gnant, University of Vienna, Vienna, Austria||“Is Endocrine Therapy of Benefit in DCIS With or Without Radiotherapy?”|
|Jay R. Harris, Dana-Farber Cancer Institute/Brigham and Women's Hospital, Boston, Mass (Writing Committee Member)||“Can We Identify Subsets of Patients Where Mastectomy Provides >10% Improvement in Local Recurrence Over Breast Conserving Surgery?”|
|Walter Jonat, University of Kiel, Kiel, Germany||“Web Tool for Decision Making”|
|Manfred Kaufmann, University of Frankfurt, Germany (Consensus Chairman and Writing Committee Member)||“Surgery and Neoadjuvant Systemic Therapy: Extent of Resection and Timing of Sentinel Node Biopsy”|
|Seema A. Khan, Northwestern University, Chicago, Ill||“Margins in Breast Conserving Surgery: What Is Optimal?”|
|Henry M. Kuerer, The University of Texas M. D. Anderson Cancer Center, Houston, Tex||“Is Axillary Dissection Always Indicated for a Positive Sentinel Node?”|
|R. Douglas Macmillan, Nottingham Breast Institute, Nottingham, United Kingdom||“Oncoplastic Surgery: Addressing the Oncological and Cosmetic Issues”|
|Gunter von Minckwitz, German Breast Group, University of Frankfurt, Frankfurt, Germany (Writing Committee Member)|
|Monica Morrow, Memorial Sloan-Kettering Cancer Center, New York, NY (Consensus Chairman and Writing Committee Member)||“Does MRI Improve the Selection or Outcome of Local Therapy?”|
|Jens Overgaard, Aarhus University Hospital, Aarhus, Denmark|
|Marie Overgaard, Aarhus University Hospital Aarhus, Denmark|
|Alain Fourquet, Institute Curie, Paris, France||“Should Post-Mastectomy Radiotherapy Be Standard for Patients With 1-3 Positive Nodes?”|
|Jean-Yves Petit, European Institute of Oncology, Milan, Italy||“Nipple Sparing Mastectomy With Intraoperative Radiotherapy”|
|Guiseppe Viale, European Institute of Oncology, University of Milan, Milan, Italy||“What Is the Optimal Technique for Sentinel Node Evaluation?”|
|Paolo Veronesi, European Institute of Oncology, University of Milan, Milan, Italy||“What Are Contraindications to Sentinel Node Biopsy in 2008?”|
|Marc van de Vijver, National Cancer Institute, Amsterdam, the Netherlands||“Molecular Profiling for the Selection of Local Therapy”|
|William C. Wood, Emory University, Atlanta, Ga||“Does DCIS Always Require Irradiation?”|
|Frederik Wenz, University of Mannheim, Mannheim, Germany||“When Can Partial Breast Irradiation Be Used Outside a Clinical Trial and Is There an Optimal Technique?”|
|Frank A. Vicini, Beaumont Hospital, Royal Oak, Mich||“Review of Trials of Excision With or Without Radiotherapy in DCIS”|
|1. What are the current contraindications to breast-conserving treatment, and what preoperative workup is needed for their identification?|
|2. What constitutes an adequate margin for breast-conserving treatment:|
|In invasive breast cancer?|
|3. What is the algorithm for management of the axilla:|
|In clinically lymph node-negative patients?|
|In clinically lymph node-positive patients?|
|4. What special considerations are necessary for surgery and radiotherapy after neoadjuvant systemic therapy?|
|5. Who should receive boost radiotherapy?|
|6. Who should receive postmastectomy radiotherapy?|
|7. What is the role of partial breast irradiation?|
|8. Are there subsets of DCIS patients not requiring radiotherapy?|
|9. How should breast reconstruction and oncologic therapy be integrated?|
|Recommendations||Levels of Evidence|
|1. Preoperative workup for BCT:||Retrospective/expert opinion|
|History, physical examination, diagnostic mammogram, ultrasound of tumor site.|
|Magnetic resonance imaging is not routinely recommended.|
|2. Definition of a negative margin:||Prospective randomized trials|
|Invasive cancer—ink not on tumor.|
|DCIS measuring 2 mm. Lesser margins do not mandate mastectomy.|
|3.What is the algorithm for management of the axilla?|
|Clinically lymph node-negative patients:||Prospective trials|
|Sentinel lymph node biopsy except in inflammatory, other T4 cancers.|
|Clinically lymph node-positive patients:||Retrospective, expert opinion|
|Image-guided fine-needle aspiration initially.|
|If positive, axillary lymph node dissection.|
|If negative, sentinel lymph node biopsy including removal of any palpable, abnormal lymph nodes.|
|Axillary lymph node dissection remains standard management for macrometastases in the sentinel lymph node.|
|4. Special considerations in neoadjuvant therapy:|
|a) The tumor site should be clipped prior to/at the initiation of systemic therapy.||Expert opinion|
|b) Surgery is necessary in patients with a clinical complete response.||Meta-analysis|
|c) The entire tumor volume does not need to be resected in tumors showing a reduction in size.|
|d) Sentinel lymph node biopsy in clinically lymph node-negative patients may be performed before or after neoadjuvant treatment. Further study is required.||Expert opinion, retrospective|
|e) Postoperative RT is recommended for all patients with BCS and for all patients with locally advanced breast cancer or initially lymph node-positive disease treated with mastectomy.||Meta-analysis|
|5. Who should receive boost radiotherapy?|
|Women aged ≤50 years with invasive cancer.|
|Women aged >50 years with high-grade tumors or lymphovascular invasion.|
|Lesser benefit in older women without high-risk features, should be considered individually.|
|6. Who should receive postmastectomy radiotherapy?|
|Patients with involvement of:||Meta-analysis randomized trial|
|≥4 axillary lymph nodes.|
|T3 tumors, any lymph node involvement.|
|No consensus on T1-2 tumors, 1-3 lymph nodes involved.|
|7. What is the role of partial breast irradiation?|
|Partial breast irradiation is not indicated outside of a trial at this time.||Evidence to support lacking|
|8. Integration of reconstruction and oncologic therapy:||Expert opinion|
|Skin sparing is contraindicated in inflammatory cancer.|
|The integration of reconstruction and PMRT should be addressed in a multidisciplinary setting.|
|Data on the effect of oncoplastic procedures on local control (ability to perform targeted re-excision, deliver boost to tumor bed) are needed.|
|9. Subsets of DCIS patients not requiring RT:||Randomized trials, meta-analysis, prospective studies|
|Prospective studies have not identified subsets not benefiting from RT.|
|1-cm margins are not a substitute for RT.|
|The absolute benefit of RT in elderly patients with widely excised areas of low- and intermediate-grade DCIS is small. Omitting RT in this group, based on patient preference, is reasonable.|
Preoperative workup to select the type of surgery
The relationship between the size of the tumor and the breast determines the possibility of breast-conserving surgery. Physical examination, diagnostic mammography, and an ultrasound of the tumor site provide reliable information on the extent of the tumor.4 Core needle biopsy distinguishes invasive from in situ disease, allows determination of tumor type and grade, and is the preferred method of diagnosis for palpable and nonpalpable lesions. Determination of estrogen and progesterone receptor and HER-2 status is necessary for planning neoadjuvant and adjuvant therapy. It is also helpful for estimating the risk of ipsilateral breast tumor recurrence (IBTR) and contralateral cancer, and may aid in assessing the need for postmastectomy RT.5 Magnetic resonance imaging (MRI) has the highest sensitivity for cancer detection, but is associated with a substantial false-positive rate. The frequency of ipsilateral and/or contralateral disease found by MRI is 2× to 4× higher than the frequency of IBTR or metachronous contralateral cancer in patients treated without MRI. Furthermore, MRI significantly increases the mastectomy rate, introduces delays in treatment, and increases costs substantially.6, 7 MRI has not been shown to increase the likelihood of negative margins, decrease unsuccessful attempts at breast-conserving surgery, or decrease the frequency of IBTR,8, 9 and the panel recommended against its routine use.
Indications and requirements for breast-conserving surgery
Breast conservation therapy (BCT) is an appropriate method of primary therapy for the majority of women with stage I and II breast cancer. BCT provides long-term survival rates equivalent to those of total mastectomy while preserving the breast.2 The risk of IBTR after BCT has decreased over time with improvements in mammography and margin assessment, and (mainly) because of the favorable interaction of adjuvant systemic therapy with RT. BCT requires the ability to deliver full-dose RT. The risk for IBTR in the absence of RT has the potential to negatively impact survival.2
Careful attention to patient selection and the technical details of surgery and RT are important, because IBTR rates in excess of 10% at 5 years are associated with increased breast cancer mortality at 15 years.2 Current contraindications to BCT are listed in Table 4. Multicentric cancer has been considered a contraindication to BCT. Recent studies suggest low rates of IBTR in selected patients (<3 lesions, no extensive intraductal component, all clinical and radiographic disease excised to clear margins).10 The Panel felt this accumulating evidence provides justification for prospective studies to determine whether multicentricity should remain an absolute contraindication to BCT.
|Disease too extensive for breast conservation||A large tumor in a small breast. Neoadjuvant therapy recommended in this circumstance if the patient desires breast conservation.|
|Persistent positive margins for invasive or intraductal disease after adequate attempts at resection.|
|Diffuse, malignant-appearing calcifications.|
|Inability to deliver radiotherapy||Previous high-dose irradiation for breast cancer, mantle irradiation for Hodgkin lymphoma, or irradiation for lung cancer.|
|First- and second-trimester pregnancy.a|
|Predictable poor cosmetic outcome||Collagen vascular disease (scleroderma or systemic lupus erythematosus).|
|Preference of patient for mastectomy||Psychological, social, or other reasons might lead to this decision after adequate information and education are provided.|
Successful BCT requires margins clear of invasive and in situ tumor and acceptable cosmetic results. There is no universal definition of what constitutes a negative microscopic margin. The only definition used in the prospective randomized trials was tumor cells not touching ink. Local control rates of 95% at 10 years are reported in patients receiving systemic therapy when this definition is used.11 Although a large number of studies show higher rates of IBTR with positive margins, we lack consistent evidence that margins wider than tumor not touching ink decrease IBTR.12 Routinely requiring larger margins may lead to more re-excisions and mastectomies and to increased psychological and financial burdens. Larger margins may be considered in selected cases based on clinical circumstances, but are not routinely warranted. Adequate pathologic examination of the surgical specimen requires specimen orientation and inking. In general, invasive or intraductal cancer at an inked surface necessitates a re-excision. The impact of focally (eg, <1 histological field at ×4 magnification) involved margins on the risk of IBTR is uncertain. Lobular carcinoma in situ at the margin is not considered an indication for further surgery.13 The Panel endorses the adoption of “tumor not touching ink” as the standard definition of an adequate negative margin in patients with invasive cancer.
Evaluation of axillary lymph nodes
Sentinel lymph node biopsy (SNB) has become the standard approach to axillary evaluation in clinically lymph node-negative patients. Ultrasound-guided fine-needle aspiration cytology is the preferred initial approach for lymph nodes suspicious clinically or on ultrasound. Inflammatory cancer is 1 of the few absolute contraindications to SNB. The accuracy of SNB for other patients with T4 breast cancer is uncertain, and the Panel does not recommend SNB for T4 tumors. Pregnancy is not a contraindication to mapping with a radioactive tracer alone; the dose of radioactivity is very low, and good outcomes have been reported in pregnant patients.14
Axillary dissection remains the standard approach for patients with sentinel lymph node macrometastases. The management of low-volume disease in the sentinel lymph node (isolated tumor cells or micrometastases) is controversial. The need for additional axillary treatment depends on the risk of additional involved lymph nodes and whether the information would influence patient management. In general, those with isolated tumor cells have the lowest risk of nonsentinel lymph node involvement.15 The use of a nomogram can be helpful to estimate risk.16, 17 In patients who will receive RT as part of BCT, radiation to an axillary field or with high tangential fields is an alternative to dissection.18
Local treatment after neoadjuvant systemic therapy
The neoadjuvant (preoperative, primary) use of cytotoxic, hormonal, and/or trastuzumab therapy effectively reduces tumor burden in the breast and the axilla without compromising survival.19, 20 The risk of local recurrence is determined by the initial clinical stage and the pathologic stage after neoadjuvant therapy.21 Neoadjuvant therapy is indicated in patients with inoperable tumors or if BCT is desired by patients with large tumors otherwise requiring mastectomy. Patients with multicentric tumors generally will not become candidates for BCT with this approach. Initial multidisciplinary evaluation is important. The tumor site should be marked before treatment (eg, by clipping) to allow tumor localization at surgery. Multiple imaging studies are not needed during treatment unless there is concern about disease progression. Surgery is necessary in all patients, even those with a complete clinical response. Any residual palpable or imaging-detected lesions should be removed, but the entire initial tumor volume does not need to be resected in tumors showing a reduction in size. The optimal timing of sentinel lymph node biopsy in relation to neoadjuvant systemic therapy in patients with a clinically lymph node-negative axilla is uncertain at this time, and further study is required. The identification rate of the sentinel lymph node appears to be lower after neoadjuvant therapy, but some patients may avoid axillary lymph node dissection because of downstaging.22 Ultrasonography of the axilla in clinically lymph node-negative patients is useful in identifying pathologically lymph node-positive patients at presentation. In some cases, knowledge of the pretreatment lymph nodal status is useful for radiation planning. In patients with lymph node-positive disease at presentation who become clinically lymph node-negative after treatment, axillary dissection is recommended because of the high false-negative rate of SNB in this circumstance.23 Postoperative RT is recommended for all patients treated with breast-conserving surgery and for all patients with initially lymph node-positive disease or with locally advanced disease treated with mastectomy.
The Early Breast Cancer Trialists Collaborative Group Overview indicates that as a rule of thumb, every 4 local recurrences avoided by RT prevents 1 death 15 years after diagnosis.2 Therefore, RT is expected to improve survival in subgroups where the absolute risk reduction for local recurrence is ≥10%. This group includes the majority of women with invasive breast cancer treated with breast-conserving surgery. A possible exception is older women with hormonally sensitive T1 tumors that lack unfavorable pathologic features. RT to the whole breast to a dose of 45 to 50 grays (Gy) (1.8-2.0 Gy per fraction) is the standard approach for patients treated by breast-conserving surgery, although recent studies have indicated that the use of hypofractionation schedules may yield a similar outcome in selected patients.24, 25 An additional radiation dose (boost) of 10 to 16 Gy reduces IBTR with a relative risk reduction of nearly 50% in all patients; however, the absolute benefit is greatest in women aged <50 years and in older women with high-grade tumors and/or lymphovascular invasion.26
Patients with ≥4 involved axillary lymph nodes treated by mastectomy benefit from postmastectomy RT.2 Computed tomography-based treatment planning to minimize irradiation of the heart and lungs is essential. The Early Breast Cancer Trialists Collaborative Group Overview2 and some individual trials27-30 also indicate a similar survival benefit for patients with 1 to 3 involved lymph nodes, but newer, more effective systemic treatments may improve local control in the absence of RT. There was no consensus on the use of RT in patients with 1 to 3 positive lymph nodes, with some favoring its routine use and others a selective use in subgroups felt to be at highest risk for local recurrence. This relationship requires further exploration with regard to risk groups, biologic subsets, and the type of systemic therapy used. postmastectomy RT is not routinely recommended for patients with T1 or T2 tumors and lymph node-negative disease.
RT to the axillary lymph nodes is not considered standard after a level I/II dissection, and current data are insufficient to provide specific recommendations on treatment of the internal mammary or supraclavicular and axillary apex lymph nodes.
There is uncertainty about the timing of RT and chemotherapy. In general, chemotherapy is given first, because it has a favorable impact on both local and distant recurrence. When additional diagnostic testing (ie, commercially available gene arrays) delays decisions regarding chemotherapy, or when tumor features indicate a higher risk of local than distant recurrence, initial RT should be considered.
Accelerated partial breast irradiation, although widely used outside of trials, is not a standard option in the absence of long-term follow-up data on its efficacy and safety, and is not recommended by the Panel outside of a trial. Patients currently treated with accelerated partial breast irradiation are highly selected for a low risk of IBTR and older age, and are not representative of the entire population undergoing BCT.31
Locoregional treatment of ductal carcinoma in situ
The risk of breast cancer death after any treatment of ductal carcinoma in situ (DCIS) is <5%, but the risk of local recurrence varies significantly by treatment. Mastectomy results in very low rates of local recurrence, but does not appear to improve survival compared with BCT. The contraindications to BCT listed in Table 4 apply to patients with DCIS, but neoadjuvant therapy to reduce the size of larger DCIS lesions is not an option. Half of the IBTRs seen after BCT for DCIS are invasive carcinoma, so minimizing IBTR is important. There is no consensus on what constitutes an appropriate margin for DCIS treated with excision and RT. As in invasive cancer, tumor-filled ducts not touching ink was the only margin definition specified in prospective randomized trials. Because DCIS, particularly low- and intermediate-grade lesions, may grow discontinuously within the ducts, there was consensus on the Panel that adequate negative margins for DCIS should be larger than those for invasive carcinoma. A recent meta-analysis indicates no reduction in IBTR with margins >2 mm in patients receiving RT32; therefore a 2-mm margin was endorsed by the Panel as an appropriate standard. Margins <2 mm are not an absolute indication for mastectomy, but should be considered along with other factors influencing IBTR, such as the amount of DCIS close to the margin, young patient age, and which margin is close. Anterior and posterior margins of <2 mm are not of concern if there is no residual breast tissue. All suspicious microcalcifications associated with the DCIS should be removed surgically. Mammograms with magnification views are recommended if any uncertainty regarding completeness of the resection remains after review of the specimen mammogram and should be routine for patients with large areas of calcifications. The only routine indication for SNB in DCIS is performance of a mastectomy. In patients treated with BCT, SNB might be offered in selected cases at higher risk for microinvasive disease (eg, palpable masses). Whole breast RT is recommended for the majority of patients, as it reduces the risk of local recurrence by approximately 50%, and subsets not benefiting from RT have not been prospectively identified.33, 34 The strongest predictors for IBTR are age <40 years and positive margins of resection. Exceptions to the use of RT might be elderly patients (particularly those with low-grade DCIS) with wide margins.
Retrospective, single institution studies have suggested that a 1-cm margin might obviate the need for RT in DCIS.35 Prospective attempts to duplicate single institution studies reporting low rates of local recurrence after wide excision and detailed tissue processing have been unsuccessful.36, 37 In patients with hormone receptor-positive DCIS, treatment with tamoxifen for 5 years reduces the risk of IBTR and contralateral cancer.38 The risk-benefit ratio of tamoxifen varies with age and is most favorable in premenopausal women. The use of aromatase inhibitors in DCIS is not recommended outside of a trial.
Neither breast reconstruction itself, nor the use of skin-sparing mastectomy to facilitate reconstruction, increases local recurrence rates,39 but skin-sparing mastectomy is technically more demanding to perform than conventional mastectomy. Skin sparing is contraindicated in inflammatory cancer. The integration of reconstruction and postmastectomy RT remains controversial, and is best addressed in a multidisciplinary setting. There are few data to judge the oncologic outcome of oncoplastic surgery, a term that encompasses a variety of approaches. The relationship between local control and survival mandates that oncoplastic procedures be examined for their effect on targeted re-excision, the ability to deliver a boost dose of radiotherapy to the tumor site, and the detection of IBTR. The assumption that local control is improved by more widely negative margins is not supported by current data.
Quality standards of locoregional treatment
Quality standards in various regions of the world have led to an improvement in outcomes. A prerequisite for achieving adequate quality is care for the patient by a multidisciplinary team. Multidisciplinary team presentation of all patients is advantageous if treatment is provided in different centers; in high-volume centers, it might be restricted to patients who cannot be treated within predefined protocols. The treatment team should select and monitor quality standards. As the aim of locoregional treatment is to prevent local recurrence, an IBTR rate of ≤1% per year after BCT is recommended as an indicator of good quality, and should be confirmed by audits of individual outcomes.
The recent Early Breast Cancer Trialists Collaborative Group Overview2 supports the importance of locoregional control. The goal of future research should be to further optimize and individualize locoregional treatment based on molecular or biological markers that remain to be identified. Targeted systemic therapies have a major impact on local control, and an improved understanding of the interaction between local and increasingly effective systemic therapy is needed. As improvements in imaging result in the identification of very small cancers that were previously unrecognized, and as the development of molecular markers that provide greater prognostic and predictive measures for local therapy become available, the rules governing the selection of surgical therapy and the use of RT will need to be reconsidered.
We thank the independent BANSS Foundation, Biedenkopf, Germany for financial support, and the academic research institution GBG Forschungs GmbH, NeuIsenburg, Germany for logistical support of the meeting.
CONFLICT OF INTEREST DISCLOSURES
This consensus symposium received financial support from the BANSS Foundation, a nonprofit body based in Biedenkopf an der Lahn, Germany. The founder of the BANSS Foundation, who died from breast cancer, wished to help extend the information resources available to clinicians and investigators in the field of oncology. Both the symposium and the preparation of this article were conducted in an independent and unbiased fashion.