Thresholds for therapies: highlights of the St Gallen International Expert Consensus on the primary therapy of early breast cancer 2009.

The 11(th) St Gallen (Switzerland) expert consensus meeting on the primary treatment of early breast cancer in March 2009 maintained an emphasis on targeting adjuvant systemic therapies according to subgroups defined by predictive markers. Any positive level of estrogen receptor (ER) expression is considered sufficient to justify the use of endocrine adjuvant therapy in almost all patients. Overexpression or amplification of HER2 by standard criteria is an indication for anti-HER2 therapy for all but the very lowest risk invasive tumours. The corollary is that ER and HER2 must be reliably and accurately measured. Indications for cytotoxic adjuvant therapy were refined, acknowledging the role of risk factors with the caveat that risk per se is not a target. Proliferation markers, including those identified in multigene array analyses, were recognised as important in this regard. The threshold for indication of each systemic treatment modality thus depends on different criteria which have been separately listed to clarify the therapeutic decision-making algorithm.

introduction

The 11th St Gallen conference held in March 2009, which was attended by >4800 participants from 101 countries, incorporated incremental information but proposed a radically different treatment selection algorithm for the management of early breast cancer. The more we know about the tumour types underlying the heterogeneity of the disease, the greater the opportunity to refine treatment choice. It was recognised that clinical trials are very useful for identifying effective treatments, but fall short of defining the optimal treatment of individual patients. For example, local control is crucial to improve survival on average and especially in patients at low risk, but is overwhelmed by the risk of distant metastases in patients at high risk. Similarly, while cytotoxic chemotherapy improves outcome on average among patients with endocrine-responsive disease receiving endocrine therapy, subgroups can be defined by conventional pathology and by multigene analyses in which little or no additional benefit accrues from chemotherapy. Judgements must be made in the care of individual patients of whether to use or withhold each treatment modality. It is the intention of this report to assist in the rational application of evolving knowledge in reaching these judgements.

St Gallen 2009: news and progress

New information was presented in the areas of genetics, tumour biology, experimental therapeutics, surgery, radiation oncology, and adjuvant systemic therapy. Some of this new information is summarised in Table 1. In the light of this information, a Panel of 43 experts from around the world (see Panel members listed in the appendix) again considered specific questions to arrive at recommended principles for the selection of therapies in early breast cancer.

Table 1.

Recent research findings presented at the 11th International Conference on Primary Therapy of Early Breast Cancer and their implications for patient care

Field or treatment	Status of research/implications for patient care
Epidemiology and changes in breast cancer incidence	Decrease in breast cancer incidence in some countries is a result of recent changes in the use of hormone replacement therapy in postmenopausal women [1]. Thus, the increased incidence that might be attributed to the use of estrogen and progestin preparations (induced carcinogenesis? induced progression of subclinical breast cancer?) is to be considered at least partly reversible [2].
Genetic predisposition	The well-established high-penetrance BRCA1 and BRCA2 genes continue to demonstrate multiple mutations (roughly 2000 each) which make testing technically difficult. Founder mutations in BRCA in some geographical areas make the detection of mutations easier.
	Genome-wide association studies define an increased number of genes which carry a smaller increase in risk for breast cancer, but are relatively common in the population. These genes are of little value in counselling individuals, though they are of biological interest and can potentially identify women at slightly increased risk which might justify selective screening policies as public health resources are limited [3].
	BRCA1 mutations are associated with triple-negative phenotype, which require clinical evaluation of novel therapeutic approaches including poly (ADP-ribose) polymerase inhibitors and DNA-damaging agents [4, 5].
Selective estrogen receptor modulator (SERM) chemoprevention	Five-year results of lasofoxifene [6] involving >8000 postmenopausal women with osteoporosis were presented. Two doses of lasofoxifene were studied: the higher dose (0.5 mg daily) proving more effective with a significantly reduced incidence of estrogen receptor-positive breast cancer (the primary study end point), overall breast cancer, vertebral fracture, nonvertebral fracture, stroke, and major coronary heart disease [7]. These latter features suggest an improved therapeutic ratio compared with tamoxifen prevention. In particular, there was no increase in endometrial cancer, though there was an increased incidence of venous thromboembolism, similar to that seen with tamoxifen.
Whole-genome studies	A cistrome is a concept incorporating the complete set of interacting related factors across the entire genome. Advancing technology allowing us to take a more comprehensive overview of events, both genetic and epigenetic, which influence particular pathways, such as those involved in steroid receptors. Within the steroid receptor cistrome, these studies have identified FOXA1 as an important component [8, 9].
	In experimental models, tamoxifen effectiveness requires HER2 suppression which is in turn regulated by the balance between PAX2 and AIB-1 [10].
Stem cells	Further support for the stem-cell hypothesis in breast cancer arises in preclinical studies in which a subpopulation of cells identified by aldefluor are uniquely capable of transplanting tumours in animal models and appear to have the characteristics of self-renewing stem cells [11]. Detection of such cells in clinical tissue microarrays identifies patients with a relatively poor prognosis [12].
microRNAs	MicroRNAs, particularly miR-335 and miR-206, affect metastases by blocking cell migration while miR-126 blocks cell proliferation. These microRNAs may be lost in highly metastatic cancers and this is associated with an oligogenic signature indicative of poor prognosis. The predictive potential is being investigated. Reintroduction of specific microRNAs has proved to be effective in suppressing metastases in animal models [13].
Networks in cellular systems	Evolution of cell survival mechanisms has required redundant network interactions rather than simple linear systems. This poses a more complex problem when attacking a cancer cell. Success is more likely to occur if two or more perturbations can be introduced, preferably at crucial early parts of the network [14]. An example is the epidermal growth factor receptor (EGFR) family, including HER2.
Circulating tumour cells	Circulating tumour cells have been increasingly studied as poor prognosis markers (though they are not yet ready for routine use). New technology allows the evaluation of phenotypic markers in individual circulating tumour cells and has demonstrated that these may differ from the gross characteristics of the parent tumour [15]. Thus, for example, HER2 overexpression in circulating tumour cells might justify targeted therapy even in the absence of conventional HER2 positivity of the primary tumour. This strategy is undergoing clinical investigation [16].
	Current studies are examining the possibility that some circulating tumour cells may represent breast cancer stem cells.
Angiogenesis	The benefits of current antiangiogenic treatment in metastatic disease are transitory. Drugs that target angiogenesis might, in the long run, induce angiogenesis as a rebound phenomenon and have been demonstrated in preclinical studies to induce tumour progression and metastases [17–19]. A possible mechanism for this tumour progression may be the release of increasing numbers of circulating endothelial cells following some types of chemotherapy. Importantly, this effect is not seen with metronomic chemotherapy [20].
	Long-term treatment with antiangiogenic drugs together with metronomic chemotherapy was associated with dramatic and profound reduction of vascular endothelial growth factor (VEGF) and substantial clinical response in metastatic breast cancer [21]. The type of cancer vascularisation and the extent of VEGF targeting might be a crucial strategic issue in the treatment of malignancies [22].
	Antiangiogenic treatments are under investigation in the adjuvant setting (but are not recommended for routine use outside clinical trials).
New opportunities for endocrine therapy	The mechanism of estrogen effect in cells resistant to estrogen deprivation is apoptosis, which is mediated by increased calcium influx [23]. Apoptosis is increased by G protein-coupled receptor 30 (GPR30). Which in turn can be induced by its agonist known as G-1 [24].
	Antiangiogenic agents enhance the tamoxifen effect [25].
	Cells which are resistant to this estrogen effect have high glutathione, and depletion of glutathione using buthionine sulphoximine (BSO) will restore full estrogen sensitivity [26].
Resistance to treatment by crosstalk	Further studies of the crosstalk between estrogen receptor and HER2 pathways show that each can act as resistance mechanism for the other. This logically led to studies combining antiestrogenic therapy with agents targeting receptors of the EGFR family. Examples included the combination of gefitinib with either tamoxifen or anastrozole and the combination of lapatinib with letrozole [27, 28].
Pharmacogenetics	The majority but not all studies have associated abnormalities of CYP2D6 on genetic grounds or as a result of certain antidepressant drugs with poorer outcome among patients treated with tamoxifen [29]. It has been indicated that increased tamoxifen dosage may overcome less effective metabolic conversion to endoxifen in some of these patients [30].
Novel imaging	Functional imaging using targets of the hormone receptor [31] and HER2 is under development [32].
Multigene assays	Multigene assays are widely proposed to add to the prognostic information available from classical pathological markers and in some circumstances have been shown to identify groups which do or do not benefit from the addition of chemotherapy to endocrine adjuvant therapy. Surveys of clinical practice indicate that the information obtained from genetic assays lead to change in treatment decisions in ∼30% of cases, mainly to avoid chemotherapy [33]. Trials to further validate this application are currently underway [34, 35]. No data are available regarding the applicability of these assays for patients with estrogen receptor-negative disease.
	Studies comparing the various genetic profiles indicate commonality in sampling groups of genes representing activation of the steroid hormone receptor pathway, the epidermal growth factor system, and markers of proliferation. While the former may be useful for specific treatment selection, the dominant prognostic information seems to reside within the proliferative marker set [36, 37].
Integrating molecular and other pathological features	Clinical, pathological, and molecular data may be integrated in more robust prognostic and predictive models. The best pathology and the most accurate assessment of established markers are key features for a choice of useful treatment, with appropriate integration of molecular assays [37] which add power to the model [38].
Surgery	Results of sentinel node biopsy after neoadjuvant chemotherapy are reliable as described in a meta-analysis [39] and supported by experience at a single institution [40].
	The definition of adequate surgical margins remains controversial with a majority of North American radiation oncologists willing to accept a margin as negative if the tumour does not extend to the inked specimen surface, while surgeons and European radiation oncologists prefer a clearance of 2–5 mm in addition to this [41]. Invasive tumour found at the inked margin is associated with increased ipsilateral breast tumour recurrence [42].
	Evidence was presented that a more generous margin was required in ductal carcinoma in situ (DCIS), perhaps reflecting the propensity of this disease to discontinuous spread [43]. Lobular carcinoma in situ (LCIS) at the margin is not regarded as an indication for reexcision [44].
	Studies to investigate the necessity of axillary dissection for patients whose sentinel node biopsy contains only micrometastatic disease (<2 mm) are underway. Meanwhile, experience from a single institution suggests that the rate of axillary recurrence remains <2% at a median follow-up of 39 months [45].
	The use of contralateral prophylactic mastectomy is clearly increasing in several series [46] though the rationale remains unclear, and evidence that this procedure improves survival is lacking [47].
Radiation therapy	Partial breast irradiation is being studied in several clinical trials but remains experimental. One application might be the treatment of patients who have already received radiation to part of the breast in the course of treatment for a previous lymphoma [48].
	Recent studies of postmastectomy radiation therapy have attempted to dissect the average survival ratio of one death prevented for every four local recurrences avoided [49]. In patients at very high risk of relapse, distant metastases predominate and local control is a less critical determinant of survival. Conversely, in low-risk cohorts, the ratio may be more favourable and has been reported to approach one death prevented for each local recurrence avoided [50].
	Accelerated partial breast irradiation is being investigated in ongoing trials, but a consensus statement from the American Society for Therapeutic Radiology and Oncology [51] provides guidance on patients who might be considered suitable for this technique outside of a study.
Endocrine therapies	Either tamoxifen or tamoxifen plus ovarian function suppression, both for the duration of 5 years, is acceptable standards for premenopausal women with endocrine-responsive disease [52, 53].
	Recent results from trials continue to support the benefit of aromatase inhibitors in postmenopausal women with receptor-positive breast cancer [54, 55], though others have questioned the extent of benefit [56]. Benefit may be particularly marked for women at higher risk of relapse. For the women at very low risk of recurrence, there appears to be little benefit from the use of aromatase inhibitors as compared with tamoxifen during the first 5 years [57]. For such patients, it may be wise to choose the best tolerated agent that maximises adherence and minimises impact on quality of life and health status. The duration of aromatase inhibitor therapy, supported by trial results, is 2–5 years [57].
HER2-targeted therapy	There is some evidence that HER2 positivity carries an adverse prognostic significance even in patients with tumours <1 cm [58], but the relationship to steroid hormone receptor status and adjuvant endocrine or cytotoxic therapies remains unclear in this group [59, 60].
Chemotherapy	There is a lack of specific predictive markers for response to individual chemotherapeutic agents. Many different regimens are used and no clear indications for a particular regimen exist. Low estrogen receptor, HER2 overexpression, and increased proliferation predict response to chemotherapy in general, rather than being agent specific [61].
Neoadjuvant systemic therapy	Preoperative cytotoxic therapy is less effective for tumours with higher levels of estrogen receptor expression [62].
Treatment of triple-negative disease	Triple-negative breast cancer is associated with an improved pathological complete response rate with neoadjuvant chemotherapy [63], but despite this there is an inferior overall survival in comparison to other breast cancer types [64]. New approaches undergoing clinical trial evaluation for treatment of triple-negative disease include new agents such as ixabepilone [65] and DNA-damaging agents such as platinum compounds, anthracyclines, and poly (ADP-ribose) polymerase (PARP) inhibitors [66].
Novel systemic treatments	Early clinical investigations are underway to evaluate several promising compounds including new anti-HER2 therapies, HSP-90 inhibitors, mTor inhibitors, anti-IGF1R mAbs, PI3K inhibitors, and antiangiogenesis drugs [67].
Follow-up after treatment for breast cancer	All the randomized trials on follow-up were conducted before availability of targeted therapies and molecular markers. A revisiting of early diagnosis of metastases to permit earlier application of targeted therapies is warranted. Intensive follow-up does not have clinical relevance. Beyond the randomized trials, new technologies including positron emission tomography scans and the detection of circulating tumour cells require further evaluation [68].
Specific news from trials	BIG 1-98: Neither the conventional sequence of tamoxifen followed by letrozole nor the reverse sequence of letrozole followed by tamoxifen proved superior to 5 years of letrozole monotherapy. Early relapses were more frequent among patients commencing treatment with tamoxifen, particularly in those at higher risk for such events. Despite substantial crossover among patients assigned tamoxifen monotherapy, the updated comparison suggested that letrozole monotherapy produced superior survival, though this did not attain conventional significance in the intent-to-treat analysis (P = 0.08) [55].
	FinHER update: Updated results of the HER2-positive component in the FinHER study confirmed the benefit of a 9-week duration treatment with trastuzumab especially if given with docetaxel (at reduced dose). Exploratory analyses suggested that the trastuzumab benefit was particularly seen among patients receiving docetaxel rather than vinorelbine during trastuzumab therapy. A prospective study is comparing this short regimen with a conventional 1-year trastuzumab regimen (SOLD trial) [69].
	HERA: Updated analyses to 4-years median follow-up confirmed the value of one year of trastuzumab in improving disease-free survival, but the overall survival analysis on an intent-to-treat basis has been complicated by substantial crossover to late use of trastuzumab in the control arm after publication in 2005 of initial study results. The 2-year treatment group remains blinded [70].

specific considerations for treatment choice

In distilling patient and tumour features to reach patient treatment decisions, the Panel has adopted a fundamentally different approach from that used in previous consensus reports [71, 72]. Clinical decisions in systemic adjuvant therapy of early breast cancer must address three distinct questions: (i) what justifies the use of endocrine therapy, (ii) what justifies the use of anti-HER2 therapy, and (iii) what justifies the use of chemotherapy. Because these decisions are based on quite separate criteria, the previous attempt to produce a single-risk categorization and a separate therapy recommendation are no longer considered appropriate. The new algorithm is summarised in Table 2. As before, the Panel recognised that adherence to therapeutic guidelines is affected by affordability of certain genetic and imaging tests and the costs of some systemic therapies in various geographic settings.

Table 2.

Thresholdsa for treatment modalities

Treatment modality	Indication	Comments
Endocrine therapy	Any ER stainingb	ER negative and PgR positive are probably artefactual [73]
Anti-HER2 therapy	ASCO/CAP HER2 positive [>30% intense and complete staining (IHC) or FISH >2.2+]b	May use clinical trial definitions
Chemotherapy
In HER2-positive disease (with anti-HER2 therapy)	Trial evidence for trastuzumab is limited to use with or following chemotherapyb	Combined endocrine therapy + anti-HER2 therapy without chemotherapy in strongly ER-positive, HER2-positive is logical but unproven
In triple-negative disease	Most patientsb,c	No proven alternative; most at elevated risk
In ER-positive, HER2-negative disease (with endocrine therapy)	Variable according to riskb	See Table 3

Most factors are continuous but a binary decision needs to be made at some level.

Patients with tumours of <1 cm in size without axillary nodal involvement and without other features indicating increased metastatic potential (e.g. vascular invasion) might not need adjuvant systemic therapy. If the tumour is, however, endocrine responsive, endocrine therapy should be considered.

Medullary carcinoma, apocrine carcinoma, and adenoid cystic carcinoma do not require chemotherapy due to low risk despite being triple negative (provided that, as is usually the case, they have no axillary node involvement and no other signs of increased metastatic risk).

ER, estrogen receptor; PgR, progesterone receptor; ASCO, American Society of Clinical Oncology; CAP, College of American Pathologists; IHC, immunohistochemistry.

endocrine therapy

The Panel recommends the inclusion of adjuvant endocrine therapy in almost all patients whose tumours show evidence of endocrine responsiveness, now defined as the presence of any detectable estrogen receptor (ER). It questioned the validity of reports of positive progesterone receptor (PgR) in the absence of ER and suggested that such cases be submitted for further pathological review. Whereas previous categories of highly endocrine responsive and incompletely endocrine responsive are not relevant to the decision to use or withhold endocrine therapy, such consideration remains important for the selection of patients with ER-positive disease to receive chemotherapy.

anti-HER2 therapy

Anti-HER2 therapy is indicated in patients with HER2-positive disease as defined by the American Society of Clinical Oncology and the College of American Pathologists (ASCO/CAP) guidelines [74]. The Panel noted that the existing trials used a slightly less restrictive definition of HER2 positivity and acknowledged that patients satisfying the inclusion criteria used in the trials might also be considered for anti-HER2 treatment.

chemotherapy

The threshold for use of cytotoxic chemotherapy is the most difficult to define. Patients receiving anti-HER2 therapy conventionally also receive chemotherapy either preceding or concurrent with the anti-HER2 treatment. Although considered logical by some of the Panel members, the use of adjuvant anti-HER2 therapy without chemotherapy remains unsupported by evidence. Chemotherapy is the mainstay of adjuvant treatment of patients with triple-negative disease who are at sufficient risk of relapse to justify its utilisation. Some rare histological types of breast cancer that fall into the category of triple negative and are diagnosed neither with axillary node involvement nor with other signs of increased metastatic potential do not require adjuvant treatment (e.g. medullary, apocrine, and adenoid cystic breast cancers). Patients with small primary tumours (pT1a pN0 and ER negative) might be spared adjuvant systemic therapy.

The threshold for recommending chemotherapy for patients with ER-positive, HER2-negative disease is particularly difficult to define. These patients include a spectrum from those at low risk [75, 76] for whom there is little evidence supporting the addition of chemotherapy to endocrine therapy and to those with high risk disease and limited ER expression where chemotherapy appears clearly justified. Table 3 summarises the characteristics which favour the use of chemotherapy, those that might justify endocrine therapy alone, and those which are not useful for making this decision. Features indicating increased risk of recurrence and thus indirectly supporting the value of adding chemotherapy to endocrine therapy in such patients include lower expression of steroid hormone receptors, grade 3 tumours, high proliferation as measured by conventional or multigene assays, and the risk factors of four or more axillary lymph nodes involved, extensive peritumoral vascular invasion, and tumour size >5 cm. Emerging data presented but not published indicate that the overall scores from multigene assays may identify patients in these high-risk categories who do not gain benefit from the addition of chemotherapy to endocrine therapy. This represents an important area of research that will likely be clarified over the next several years. Patients with high expression of ERs and PgRs (e.g. >50%), grade 1 tumours, low proliferation, negative axillary lymph nodes, no peritumoral vascular invasion, and tumour size ≤2 cm may be considered for endocrine therapy alone. We note that some features individually provide little guidance in reaching a decision to use chemotherapy. In particular, histological grade 2, intermediate scores on multigene assays, tumour size between 2 and 5 cm, and low numbers of involved lymph nodes (one to three) do not provide definitive indications to either give or withhold chemotherapy. However, if all these intermediate criteria are present, it usually tips the balance towards the use of chemotherapy. The Panel considered the available multigene assays in this context and concluded that a validated assay should be taken into account as an adjunct to high-quality pathology phenotyping if doubt about the indication for chemotherapy persists after consideration of other factors. Considerations of availability and cost determine the current usefulness of multigene assays. The Panel noted that patients with pT1a pN0 and ER-positive disease should be offered endocrine therapy alone even if features which usually indicate chemotherapy are present.

Table 3.

Chemoendocrine therapy in patients with ER-positive, HER2-negative disease

	Relative indications for chemoendocrine therapy	Factors not useful for decision	Relative indications for endocrine therapy alone
Clinicopathological features
ER and PgR	Lower ER and PgR level		Higher ER and PgR level
Histological grade	Grade 3	Grade 2	Grade 1
Proliferation	Higha	Intermediatea	Lowa
Nodes	Node positive (four or more involved nodes)	Node positive (one to three involved nodes)	Node negative
PVI	Presence of extensive PVI		Absence of extensive PVI
pT size	>5 cm	2.1–5 cm	≤2 cm
Patient preference	Use all available treatments		Avoid chemotherapy-related side-effects
Multigene assays
Gene signatureb	High score	Intermediate score	Low score

Conventional measures of proliferation include assessment of Ki67-labelling index (e.g. low, ≤15%; intermediate, 16%–30%; high, >30%) [77] and pathological description of the frequency of mitoses. The reliability of these measures will vary in different geographic settings. First-generation genetic signatures contain genes sampling the ER, HER2, and proliferative pathways [78, 79]. Meta-analysis indicates that much of the prognostic information in these signatures resides in their sampling of proliferative genes [80], but their respective total scores may be the only form in which information is provided at present and could be used in this component of assessment of relative indications for chemotherapy.

The Panel agreed that validated multigene tests, if readily available, could assist in deciding whether to add chemotherapy in cases where its use was uncertain after consideration of conventional markers.

ER, estrogen receptor; PgR, progesterone receptor; pT, pathological tumour size (i.e. size of the invasive component); PVI, peritumoral vascular invasion.

endocrine responsiveness

Based on the philosophy of defining categories according to their implications for treatment selection, the previous three categories of endocrine responsiveness have been simplified so that endocrine therapy is considered indicated if any ER staining is present in the tumour. The majority of Panellists were in favour of indicating the percentage of stained cells on pathology reports rather than merely using scores. Staining for hormone receptors of ≥50% of tumour cells was viewed as indicating highly endocrine-responsive tumours.

HER2 positivity

Two technologies are recognised for the determination of HER2 positivity. These have recently been addressed by a joint working party of the ASCO/CAP [74]. Either immunohistochemical analysis showing uniform, intense membrane staining of >30% of the tumour cells or, alternatively, determination of gene amplification by fluorescence in situ hybridisation (FISH) (ratio of HER2 gene copies to chromosome 17 centromers >2.2) or chromogenic in situ hybridisation (CISH) (more than six HER2 signals per nucleus) is sufficient to define HER2 positivity. Although the definitions used in the pivotal trials of trastuzumab were less restrictive [81-84], a substantial minority of the Panellists preferred to use 30% intense and complete staining as a threshold for recommendation of anti-HER2 therapy.

pathological evaluation of characteristics of the disease

In addition to reporting the presence and type of tumour, the Panel considered various additional pathological parameters. Markers of proliferation, and specifically Ki-67-labelling index, were considered important for the determination of prognosis and, importantly, to indicate the potential value of the addition of chemotherapy to patients with receptor-positive disease. Ki-67 specifically was not accepted as the basis for choosing aromatase inhibitors rather than tamoxifen in postmenopausal patients with receptor-positive disease [85] as further validation of findings in this regard was felt to be necessary [86]. Reporting of ER generated considerable discussion. The Panel strongly endorsed the reporting of percentage of stained cells but was evenly divided on whether other scoring methods should also be reported. PgR was considered valuable for prognosis, but less important for predicting response to treatment (e.g. tamoxifen).

The majority of the Panel considered that high grade was a sufficient indication for chemotherapy and that genomic grade could be considered as an adjunct to histological grade if readily available. Gene expression signatures are likely to indicate a prognostically relevant dichotomy (low grade versus high grade), though the implications of this observation for therapy require further study [87, 88]. uPA/PAI-1 was not accepted by a majority of the Panel as a useful prognostic factor.

In an important change from the previous St Gallen conference and after a long debate, the Panel supported the use of a validated multigene-profiling assay, if readily available, as an adjunct to high-quality phenotyping of breast cancer in cases in which the indication for adjuvant chemotherapy remained uncertain.

local and regional treatments

The aspects considered by the Panel included surgical margins, indications for sentinel node biopsy, and the role of prophylactic mastectomy. Re-excision was considered mandatory if invasive cancer or DCIS is present at the inked surgical margin, but is not required for lobular carcinoma in situ (LCIS). The Panel was divided about the need for surgical margins greater than “not on ink” in DCIS, although no detailed specific recommendation was given beside avoiding the need to insist on a large (e.g. 1 cm) free margin. The use of surgical procedures developed to allow a wide excision with satisfactory results (oncoplastic surgery) was also endorsed. The Panel considered that sentinel node biopsy is the standard of care for patients with a clinically negative axilla and that axillary node dissection could be avoided in all patients with a negative sentinel node and in selected patients with micrometastatic disease or isolated tumour cells in the sentinel node. A trend to increasing use of prophylactic contralateral mastectomy was reported, though it was acknowledged that this procedure was not associated with any proven survival advantage.

Radiation therapy after local excision of DCIS was considered to be standard by the Panel members, though most members considered that it could be avoided in elderly patients and those with low-grade DCIS and clearly negative margins. For patients with invasive cancer, postmastectomy radiation therapy was indicated for those with four or more involved axillary lymph nodes, but indications for its use in patients with one to three nodes were considered more restricted and particularly applicable for young patients and those with other poor prognostic features. The majority of the Panel considered that accelerated whole-breast radiation after conservative surgery was an acceptable option for patients aged ≥60 with cancers with favourable patterns, but that partial breast radiation should still be considered experimental. The Panel considered that endocrine therapy without radiation might be considered in elderly patients with small tumours, clinically node-negative and -positive ERs.

adjuvant systemic therapies

The Panel considered targeted therapies against the steroid hormone receptors and overexpressed HER2 as of prime importance. In patients whose tumours lack these targets or in those at higher risk despite the presence of steroid hormone receptors, the use of chemotherapy requires consideration as set out in Tables 2 and 3.

endocrine therapy for premenopausal patients

The Panel accepted either tamoxifen or tamoxifen plus ovarian function suppression as standard endocrine therapies in this group. Ovarian function suppression alone or ovarian ablation was considered a possibility only in extraordinary circumstances. Aromatase inhibitors alone are contraindicated in premenopausal patients. In case tamoxifen is contraindicated, aromatase inhibitors may be administered to premenopausal women together with ovarian function suppression. Verification of ovarian function suppression to postmenopausal levels is important also in patients under the age of 60 who are receiving aromatase inhibitors.

Pharmacogenetic determination of tamoxifen metabolism status as influenced by CYP2D6 was not considered ready for routine application in selecting patients for tamoxifen therapy by the majority of the Panellists.

endocrine therapy in postmenopausal patients

A majority of the Panel considered that an aromatase inhibitor should form part of standard endocrine therapy for postmenopausal women with receptor-positive breast cancer, though acknowledging that there were certain patients for whom tamoxifen alone can be considered adequate. There was division about the proper duration of treatment with aromatase inhibitors, though it was pointed out that safety data beyond 5 years are not yet available. The majority of the Panel preferred aromatase inhibitors as up-front endocrine treatment particularly in patients at higher risk of early relapse.

anti-HER2 therapy

Updated results from two of the trastuzumab trials were presented continuing to demonstrate the value of this therapy for patients with HER2-positive disease. The FinHER trial evaluated a short course of trastuzumab, which is currently being compared with a conventional 1-year duration. Meanwhile, the standard duration of trastuzumab therapy remains 1 year. The Panel noted that no results are yet available from the 2-year trastuzumab group in the HERA trial. Interestingly, a majority of the Panel was prepared, for selected women, to contemplate trastuzumab with endocrine therapy but without chemotherapy despite the absence of clinical trial evidence to support this approach. Finally, the limited evidence of increased risk among patients with HER2-positive tumours <1 cm in size without axillary nodal involvement does not allow definitive recommendation regarding anti-HER2 therapy in this group.

adjuvant chemotherapy

Two situations were recognised in which the decision to use adjuvant chemotherapy was relatively clear-cut. First, adjuvant systemic therapy for patients with triple-negative disease is essentially limited to chemotherapy, and most such patients are at sufficient risk to justify this treatment. Secondly, as noted above, chemotherapy is conventionally given with or preceding trastuzumab for patients with HER2-positive invasive breast cancer. The remaining patients—those with ER-positive, HER2-negative disease—are the group in whom decisions about adjuvant chemotherapy are most difficult (Table 3). The Panel recognised that patients whose tumours contained high levels of ER derived less benefit from addition of chemotherapy to endocrine therapy. There was no agreement about the definition of a standard chemotherapy regimen for any disease subset. Taxane-containing regimens were discussed and combinations containing docetaxel and cyclophosphamide as well as dose-dense doxorubicin and cyclophosphamide followed by paclitaxel were viewed as standard therapies among several other regimens.

neoadjuvant systemic therapy

Neoadjuvant systemic therapy was considered justified primarily to enhance the possibility of breast-conserving surgery. If indicated, the majority of the Panel considered that the neoadjuvant chemotherapy regimen should include both a taxane and an anthracycline and (for HER2-positive disease) an anti-HER2 drug. Thus, the choice of a regimen for adjuvant or neoadjuvant chemotherapy might be made using similar criteria. Neoadjuvant endocrine therapy without chemotherapy was considered reasonable for postmenopausal patients with strongly receptor-positive disease. If used, such treatment should be considered for a duration of 5–8 months or until maximum tumour response.

preservation of fertility

Pregnancy after diagnosis of breast cancer has not been shown to negatively impact prognosis. Women should be counselled about options for preserving fertility. The Panel did not consider that any currently available methods for preservation of fertility following chemotherapy were of proven value, though gonadotropin-releasing hormone agonists are used occasionally. These are being tested in an ongoing clinical trial for women with endocrine nonresponsive disease who are receiving alkylating agents. Cryoconservation and retransplantation of ovarian tissue are also experimental.

use of bisphosphonates

Emerging information on bone protection from demineralisation and tumour by bisphosphonates was viewed as interesting, but the Panel did not consider that routine use of bisphosphonates was indicated for women with normal bone health receiving adjuvant endocrine therapy.

male breast cancer

The Panel considered that adjuvant tamoxifen was standard therapy and did not endorse the use of adjuvant aromatase inhibitors in men with breast cancer.

commentary

The present report proposes a new approach to the separate selection of each treatment modality according to its most relevant indications. We look forward to future studies more accurately defining the value of various high-throughput technologies in assessing the level of risk and likelihood of response to specific therapies. Meanwhile, careful application of the presently available therapies described in this report offers great value to women with early breast cancer.