Review
Predictors of axillary lymph node metastasis in breast cancer: A systematic review

https://doi.org/10.1016/j.ejso.2006.09.003Get rights and content

Abstract

Aims

To review the established and emerging techniques in axillary lymph node prediction and explore their potential impact on clinical practice. To reliably identify patients in whom axillary lymph node surgery, including SLNB, can be safely omitted.

Methods

Searches of PubMed were made using the search terms “axilla” (or “axillary”), “lymph”, “node” and “predictor” (or “prediction”). Articles from abstracts and reports from meetings were included only when they related directly to previously published work.

Findings

There are numerous studies in which the predictive utility of biomarkers as determinants of axillary lymph node status have been investigated. Few of these have specifically addressed the attributes of the primary tumour which could offer much potential for the prediction of tumour metastasis to the axillary lymph nodes.

Conclusions

Currently, no single marker is sufficiently accurate to obviate the need for formal axillary staging using SLNB or axillary clearance.

Introduction

Breast cancer is an enigmatic disease with an unpredictable natural history and has been a fertile soil for the development of hypothetical models of the natural history of the disease process with therapeutic consequences.1 The dominant model for breast cancer in the 19th century was a mechanistic model describing the disease as a local phenomenon within the breast spreading centrifugally along lymphatics to first and then second echelon nodes, with progressive systemic spread. Radical local surgery, the Halstead radical mastectomy, was a therapeutic consequence of this model. In the late 1960s, this model was replaced by a biological model described by Fisher,2 in which disease outcome is pre-determined by the extent of micrometastatic dissemination, early on in the natural history of the disease. Depending on the assessment of recognised prognostic markers, at diagnosis,3 prognosis is now believed to be either ‘favourable’ or ‘unfavourable’ rather than disease being regarded as ‘early’ or ‘late’. The National Surgical Adjuvant Breast Project B04 trial4 reported that axillary lymph node dissection was important for achieving local control and obtaining prognostic information, but that prophylactic removal of the lymph nodes did not improve survival. This supports the biological model rather than the mechanical model of stepwise spread. The therapeutic consequence of this model has been the development of adjuvant systemic therapy using cytotoxic drugs or endocrine manipulation.

In keeping with this model, surgical treatment of breast cancer has evolved substantially in the last three decades away from extensive surgical resection towards a more tailored and conservative approach to the breast. Two randomised controlled trials with over 20 years of follow-up demonstrated that breast conserving surgery followed by radiotherapy is as effective as mastectomy with no detrimental effect on patient survival.5, 6 However, there are some flaws in the biological model of disease such as the unpredictability of cancer spread and the survival advantage observed with post-mastectomy chest wall radiotherapy.7 These inconsistencies could be explained by using non-linear mathematical models to better predict patient prognosis.1

As the proportion of breast cancers detected via mammographic screening increases, so the incidence of axillary node involvement at diagnosis decreases and this undoubtedly changes the risk/benefit of routine axillary surgery.8 The detection of these small, sometimes sub-clinical, breast carcinomas further highlights the need for more conservative surgical procedures. The implementation of mammographic screening has also resulted in the diagnosis of DCIS in approximately 20% of all screen detected breast cancers9 but only 1% of DCIS patients have been found to have axillary nodal involvement following axillary dissection.10 Hence, routine axillary surgery is unjustified given the low yield and morbidity of such a procedure.

The role for axillary lymph node dissection is also likely to face further scrutiny with regard to its use in the selection of adjuvant treatments. Systemic therapy is increasingly being recommended to women with node negative breast cancer, this recommendation is based on the evidence of equivalent proportional survival benefit, independent of axillary lymph node status.11, 12

More recently, with the introduction of sentinel node biopsy, there has also been a move towards a more tailored and conservative approach to the axilla.

The sentinel lymph node (SLN) refers to the first regional node or nodes to drain the primary tumour. SLN biopsy (SLNB) is a minimally invasive technique (Fig. 1a and b) and its associated morbidity is less than that of axillary dissection in terms of post-operative pain, arm mobility, lymphoedema, paraesthesia and cosmesis.13 It is still an invasive procedure and complications include pain, paraesthesia and arm swelling up to 24 months after SLNB surgery.13

SLNB is highly accurate with a sensitivity of 95% and a specificity approaching 100%.13 Patients with a negative SLN biopsy are not offered further axillary surgery14, 15, 16, 17, 18 whereas patients with macrometastases (>2 mm cancer focus) are offered axillary node clearance (or radiotherapy). The axillary management of patients with micrometastasis (0.2–2 mm cancer focus) or isolated tumour cells is controversial19 with many centres offering axillary node clearance. If we accept the biological model of disease in which prognosis is pre-determined at diagnosis by the extent of micrometastatic spread, a fundamental question arises as to why we are continuing to treat sub-clinical axillary disease differently from micrometastatic disease elsewhere in the body. It is therefore important to consider how we can better select patients for the most appropriate axillary procedure and if sentinel node biopsy and axillary node clearance can be safely omitted altogether, in selected cases.20, 21

SLNB does have some drawbacks which need to be considered. SLNB has not been shown to provide the therapeutic benefit of axillary lymph node clearance or radiotherapy and there is little data on long-term morbidity and loco-regional control. SLN biopsy carries a significant false negative rate of 5–10% and such under-staging of the axilla could have considerable impact on decisions for further surgery and adjuvant treatment.22 Over 65–70% of all SLN biopsies are negative23 and these benign SLN biopsies are therefore potentially unnecessary. In cases where the SLN is involved, axillary clearance fails to reveal any further involved lymph nodes in more than half of all patients.23 Therefore, some women with a positive SLN could be spared further axillary dissection potentially without compromising either local control or survival.24 Prognostic tools are required to better select patients for axillary surgery and spare the majority of women unnecessary surgery.

It is important from the outset, to make a distinction between prognostic and predictive factors. Prognostic factors are clinical, pathologic or biological features of cancer patients that predict clinical outcome such as disease-free or overall survival. Predictive factors are clinical, pathologic or biological features that estimate the likelihood of axillary lymph node metastasis. Developments in scientific methodology have generated an ever-increasing number of markers for primary and metastatic disease but none of these are in widespread clinical use.

Using biological parameters of the primary tumour to estimate the risk of axillary lymph node involvement could provide a rationale for proceeding to SLNB or axillary sampling. Predictors may be of particular clinical utility in cases where the SLNB is equivocal (contains micro-metastases) or negative, offering a strategy to reduce the false negative rate. They may also offer an alternative to invasive techniques in prognostication and planning of adjuvant treatment. The benefit of accurate predictors of lymph node status is likely to be greatest in those patients who are axillary lymph node negative at the time of diagnosis, these individuals could be spared the risk, cost and distress of a negative axillary procedure. Predictors may also find utility in situations where complete lymph node sampling is not possible, for example, in the context of neo-adjuvant chemotherapy which affects the ability to perform accurate surgical staging and situations where co-morbidity, logistical, or financial considerations prevent axillary dissection.

Predictors in the SLN indicative of further positive axillary lymph nodes have recently been reported. Callejo et al.25 identified lymphovascular invasion, sentinel lymph node metastasis > 2 mm and the presence of extranodal spread as statistically significant predictors of the metastatic involvement of non-sentinel lymph nodes. In addition, tumour size, the number of examined sentinel lymph nodes and proportion of involved SLN have been correlated with non-sentinel metastases.26, 27, 28 The biological significance of SLN micro-metastases and their utility in guiding the management of the axilla and systemic adjuvant therapy has been the subject of much debate. Occult micro-metastases have not been found to be significantly associated with loco-regional recurrence29 or prognosis.30

The use of attributes of the primary tumour for prognostication, planning and optimisation of adjuvant treatment is currently under study in the MINDACT and EORTC 10994 trials.31 It is conceivable that we may eventually look to the primary tumour, rather than the axillary lymph nodes, to provide guidance for adjuvant local and systemic treatments. We reviewed the established and emerging techniques in axillary lymph node prediction and explored their potential impact on clinical practice. We evaluated if these could be used to reliably identify patients in whom axillary lymph node surgery, including SLNB, can be safely omitted.

Section snippets

Search strategy and selection criteria

Data for this review were identified by searches of PubMed and references from relevant articles using the search terms “axilla” (or “axillary”), “lymph”, “node” and “predictor” (or “prediction”). Articles included were written in English and published between 1966 and July 2006. Articles from abstracts and reports from meetings were included only when they related directly to previously published work. The abstracts of publications found were screened and selected by consensus by the first

Results

The literature search yielded 290 articles of which 30 articles fulfilled the selection criteria. These included 66 separate analyses (8 univariate and 58 multivariate) on 27 predictive parameters. Two additional very relevant studies are shown in the Table 1 but provided incomplete data for our analysis.

The attributes of the primary tumour have been categorised into clinical, radiological, pathological and molecular. Results are reviewed below and are summarised in Table 1a, Table 1b.

Discussion

There are numerous studies in which the predictive utility of biomarkers as determinants of axillary lymph node status have been investigated, few of these have specifically addressed the attributes of the primary tumour which could offer much potential for the prediction of tumour metastasis to the axillary lymph nodes. In order for these markers to be of clinical utility, the methodology employed on post-resection tumour samples must be applicable to pre-operative biopsy material. Many of the

Conclusion and recommendation for clinical practice

Currently, no single marker or combination of markers is sufficiently accurate to obviate the need for formal axillary staging. Markers cannot be used to justify withholding SLNB from selected patients with very favourable prognostic features or in omitting SLNB (and proceeding to axillary node clearance) in patients with unfavourable prognostic features.

Acknowledgements

The authors would like to thank Cancer Research UK for their generous contribution to this review by way of the Core Skills & Training Bursaries Fund. We would also like to thank the Academy of Medical Sciences and The Health Foundation for their generous support of this work through the Clinician Scientist Scheme.

Conflicts of interest: None declared.

References (83)

  • M. Noguchi et al.

    Clinical and biological prediction of axillary and internal mammary lymph node metastases in breast cancer

    Surg Oncol

    (1993)
  • B. Fisher

    Seminars of Bernard Fisher 1960—nature of cancer as systemic disease?

    Bull Soc Int Chir

    (1972)
  • B. Fisher et al.

    Ten year follow-up results of patients with carcinoma of the breast in a co-operative clinical trial evaluating surgical adjuvant chemotherapy

    Surg Gynecol Obstet

    (1975)
  • N. Wolmark et al.

    Surgery in the primary treatment of breast cancer

    Breast Cancer Res Treat

    (1981)
  • U. Veronesi et al.

    Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer

    N Engl J Med

    (2002)
  • B. Fisher et al.

    Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer

    N Engl J Med

    (2002)
  • M. Overgaard et al.

    Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish breast cancer cooperative group 82b trial

    N Engl J Med

    (1997)
  • B. Cady et al.

    The new era in breast cancer. invasion, size, and nodal involvement dramatically decreasing as a result of mammographic screening

    Arch Surg

    (1996)
  • I.S. Fentiman

    The dilemma of in situ carcinoma of the breast

    Int J Clin Pract

    (2001)
  • M.J. Silverstein et al.

    Axillary lymph node dissection for T1a breast carcinoma. Is it indicated?

    Cancer

    (1994)
  • Polychemotherapy for early breast cancer: an overview of the randomised trials. Early Breast Cancer Trialists' Collaborative Group

    Lancet

    (1998)
  • Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy. 133 randomised trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Early Breast Cancer Trialists' Collaborative Group

    Lancet

    (1992)
  • U. Veronesi et al.

    A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer

    N Engl J Med

    (2003)
  • A.E. Giuliano et al.

    Lymphatic mapping and sentinel lymphadenectomy for breast cancer

    Ann Surg

    (1994)
  • A.E. Giuliano et al.

    Sentinel lymphadenectomy in breast cancer

    J Clin Oncol

    (1997)
  • U. Veronesi et al.

    Sentinel Lymph Node Biopsy and Axillary dissection in breast cancer: results in a large series

    J Natl Cancer Inst

    (1999)
  • G.F. Schwartz et al.

    Proceedings of the consensus conference on the role of sentinel lymph node biopsy in carcinoma of the breast, April 19–22, 2001, Philadelphia, Pennsylvania

    Cancer

    (2002)
  • G. Cserni et al.

    Meta-analysis of non-sentinel node metastases associated with micrometastatic sentinel nodes in breast cancer

    Br J Surg

    (2004)
  • P.M. Ravdin et al.

    Prediction of axillary lymph node status in breast cancer patients by use of prognostic indicators

    J Natl Cancer Inst

    (1994)
  • R.E. Mansel et al.

    First results of the randomized multicenter ALMANAC trial

    J Clin Oncol

    (2004)
  • G. Viale et al.

    Predicting the status of axillary sentinel lymph nodes in 4351 patients with invasive breast carcinoma treated in a single institution

    Cancer

    (2005)
  • L. Kocsis et al.

    Use and limitations of a nomogram predicting the likelihood of non-sentinel node involvement after a positive sentinel node biopsy in breast cancer patients

    Am Surg

    (2004)
  • I.P. Callejo et al.

    Predictors of positive axillary lymph nodes in breast cancer patients with metastatic sentinel lymph node

    Clin Transl Oncol

    (2005)
  • G. Cserni et al.

    Axillary sentinel node and tumour-related factors associated with non-sentinel node involvement in breast cancer

    Jpn J Clin Oncol

    (2004)
  • F.J. Fleming et al.

    Factors affecting metastases to non-sentinel lymph nodes in breast cancer

    J Clin Pathol

    (2004)
  • W.K. Hung et al.

    Non-sentinel lymph node metastases in breast cancer patients with metastatic sentinel nodes

    ANZ J Surg

    (2005)
  • I. Langer et al.

    Axillary recurrence rate in breast cancer patients with negative sentinel lymph node (SLN) or SLN micrometastases: prospective analysis of 150 patients after SLN biopsy

    Ann Surg

    (2005)
  • A. Khan et al.

    Comprehensive axillary evaluation in neoadjuvant chemotherapy patients with ultrasonography and sentinel lymph node biopsy

    Ann Surg Oncol

    (2005)
  • K. Anan et al.

    Axillary lymph node metastases in patients with small carcinomas of the breast: Is accurate prediction possible?

    Eur J Surg

    (2000)
  • I.A. Olivotto et al.

    Prediction of axillary lymph node involvement of women with invasive breast carcinoma: a multivariate analysis

    Cancer

    (1998)
  • J. Ahlgren et al.

    Prediction of axillary lymph node metastases in a screened breast cancer population. South-east Sweden Breast Cancer Group

    Acta Oncol

    (1994)
  • Cited by (64)

    • The variability of aggressiveness of grade 1 breast cancer

      2023, Journal of Gynecology Obstetrics and Human Reproduction
    • The Immune Response in Nonmetastatic Axillary Lymph Nodes Is Associated with the Presence of Axillary Metastasis and Breast Cancer Patient Outcome

      2020, American Journal of Pathology
      Citation Excerpt :

      However, a limitation of our study is that using TMAs is more difficult for studying other histologic or stromal variables, as noted by Grigoriadis et al.33 The use of well-established and new pathologic factors, together with the immune and stromal variables, could enable the more accurate prediction of BC patients. In our study, the main classic factors associated with the presence of ALN+ at diagnosis in the univariate analyses coincide with those highlighted by other studies (ie, primary tumor size, histologic grade, LVI, and PNI) that are used in current clinical practice.32,34 Our results suggest that LVI masks other variables associated with the circumstance of having ALN+ at diagnosis.

    View all citing articles on Scopus
    View full text