Letters

Sentinel node biopsy in breast cancer

BMJ 1999; 318 doi: https://doi.org/10.1136/bmj.318.7183.599 (Published 27 February 1999) Cite this as: BMJ 1999;318:599

Effect on patients must be considered

  1. Hazel Thornton, Chairman, Consumers' Advisory Group for Clinical Trials (Hazelcagct{at}aol.com)
  1. Saionara, Rowhedge, Colchester CO5 7EA
  2. Sarcoma and Melanoma Unit, Royal Marsden NHS Trust, London SW3 6JJ
  3. Guy's and St Thomas's Breast Unit, Guy's and St Thomas's Hospitals, London
  4. Academic Office, Edinburgh Breast Unit, Western General Hospital, Edinburgh, EH4 2XU

    EDITOR—Careful consideration needs to be given to the implications for breast cancer patients from the use of the unevaluated technique of sentinel node biopsy.1 Ideally, clinical trials to refine this technique should have been done before its introduction into routine clinical practice by enthusiastic surgeons. Training courses and conferences on this topic must be timed to avoid encouraging premature introduction and wasting resources.

    The possibility that sentinel node biopsy would avoid clearance of the axilla, with its attendant problems, could be seductive to both surgeons and patients. The technique holds promise of clearer prognosis and swifter assessment with minimum invasiveness. What then might the drawbacks be for patients? Consideration of the best total outcome for the patient must always be the overriding objective.

    It might be useful to draw a parallel between the introduction of mammographic screening, which resulted in closer scrutiny of breast tissue, and the use of sentinel node biopsy. Screening caused the label “breast cancer” to be applied to borderline cases so that many women with non-invasive conditions which would never cause them problems in their lifetime carried the “cancer” label with all its drawbacks, emotional and financial. By the same token, closer scrutiny of the sentinel node will cause an upstaging of the disease. Techniques such as step sections, immunostaining, and polymerase chain reaction (which can detect 1in 10000 cancer cells) “metastatic” is applied more readily, with enormous implications for the women given that label.

    Foucar commented on the surprising docility of patients about the pathologist's monopoly on diagnostic terminology that links objective histopathological observations to clinical interventions.2 Classifications resulting from the use of sentinel node biopsy will not only give choice of therapy but change attitudes of mind. We will therefore need to implement Foucar's recommendation to develop a new classification so that we do not transmit “more fear than knowledge into the clinical arena.”

    References

    Arguments for node biopsy are weak

    1. Meirion Thomas, Consultant surgeon
    1. Saionara, Rowhedge, Colchester CO5 7EA
    2. Sarcoma and Melanoma Unit, Royal Marsden NHS Trust, London SW3 6JJ
    3. Guy's and St Thomas's Breast Unit, Guy's and St Thomas's Hospitals, London
    4. Academic Office, Edinburgh Breast Unit, Western General Hospital, Edinburgh, EH4 2XU

      EDITOR—Sentinel node biopsy was developed in the treatment of malignant melanoma for two good reasons—namely, reduced morbidity and possible enhanced survival. When elective regional node dissection was common practice, the technique offered a method of selecting patients for regional node dissection. Because most melanomas occur in the leg, that meant reducing the incidence of lymphoedema, which occurs after inguinal node dissection.

      Two controlled trials have shown no advantage from elective regional node dissection,1 but many large uncontrolled series have claimed a survival advantage for patients with micrometastatic nodal disease.2The concern was that the two controlled trials had not been large enough to identify a subset of patients with tumours within a given range of Breslow thickness who might benefit from regional node dissection.

      However, this argument for sentinel node biopsy does not translate comfortably to the treatment of breast cancer, where the possible benefit relates entirely to post-surgical morbidity. But what morbidity? The incidence of moderate and severe lymphoedema after regional node dissection of the axilla is minimal.3 Furthermore, an experienced surgeon can perform axillary node dissection in about the same time as it takes to perform a sentinel node biopsy. Those who justify sentinel node biopsy for non-palpable screen detected carcinoma because the incidence of axillary node metastases is so low should instead consider a selective policy for node dissection. Do patients with small grade I tumours and those with invasive duct carcinomas (say less than 1cm) need axillary clearance? Could observation alone be used or the axilla included in the field of irradiation?

      An attractive argument for sentinel node biopsy is the concept of upstaging node negative patients as a result of immunohistochemical analysis4 of the sentinel node, but that technique could easily be introduced to evaluate negative nodes after axillary clearance. However, we do not know the effect of micrometastatic disease on treatment and prognosis, and we are many decades away from the results of clinical trials that will determine the effect of adjuvant therapy on patients with positive sentinel nodes.

      If the argument for sentinel node biopsy in breast cancer is so weak, why is there such enthusiasm? I could be uncharitable and suggest that this research will keep some academic breast units in publications for years. More ominously, the fire is undoubtedly fuelled by the companies who produce the (very expensive) gamma probes and who will currently willingly sponsor surgical and breast meetings.

      References

      Value is already proved

      1. N Beechey-Newman, Consultant surgeon,
      2. S Fentiman,, Professor of surgical oncology,
      3. A E Young, Consultant surgeon
      1. Saionara, Rowhedge, Colchester CO5 7EA
      2. Sarcoma and Melanoma Unit, Royal Marsden NHS Trust, London SW3 6JJ
      3. Guy's and St Thomas's Breast Unit, Guy's and St Thomas's Hospitals, London
      4. Academic Office, Edinburgh Breast Unit, Western General Hospital, Edinburgh, EH4 2XU

        EDITOR—Dixon's editorial on sentinel node biopsy in breast cancer summarises the development and expected benefits of a powerful and relatively new technique.1

        Although some important practical aspects need to be addressed, such as the method of achieving accurate intraoperative histological evaluation, extensive evidence now exists on the principal part of the technique.2 3 a large number of well conducted studies have compared histology after sentinel node biopsy and axillary lymphadenectomy in the same patients. Many other studies are nearing completion. Dixon is therefore unnecessarily cautious in stating that it is not yet time for “trials comparing sentinel node biopsy with standard techniques of assessing axillary node disease.” These trials have already been done, and the results show that sentinel node biopsy gives an accurate indication of axillary node status in 92.3-100% of cases.4

        Sentinel node biopsy is a robust technique that can provide accurate results by whichever method is used. Success often depends more on the skill and experience of the surgeon than on the method used. With accuracy rates for different techniques all around 95%, clinical trials comparing the techniques are neither necessary nor possible. What is needed is arrangements within each unit to validate the techniques in their own hands before using them as the sole means of assessing the axilla.

        Further studies will be helpful in answering some of the remaining ancillary issues, but it would be a disadvantage if the momentum and clinical interest that have already developed were held up as a result of ‘s editorial.

        References

        Author's reply

        1. Michael J Dixon, Honorary senior lecturer in surgery (jmd{at}wght.demon.co.uk)
        1. Saionara, Rowhedge, Colchester CO5 7EA
        2. Sarcoma and Melanoma Unit, Royal Marsden NHS Trust, London SW3 6JJ
        3. Guy's and St Thomas's Breast Unit, Guy's and St Thomas's Hospitals, London
        4. Academic Office, Edinburgh Breast Unit, Western General Hospital, Edinburgh, EH4 2XU

          EDITOR—Thornton supports my view that sentinel node biopsy should be refined in clinical trials before it is introduced into routine clinical practice. A two phase trial has been funded in the United Kingdom by the Medical Research Council; the initial phase is to show that the centres involved in the study can consistently identify sentinel nodes, and the randomised phase will compare sentinel node biopsy with standard axillary staging techniques. Within this and other ongoing studies1 it will be possible to assess the importance of metastatic disease in lymph nodes identified only by immunohistochemical techniques.

          Her comments on screening deserve comment. The recent update of the national surgical adjuvant breast project trial of ductal carcinoma in situ reports that after apparently complete excision of ductal carcinoma in situ over one quarter of women develop recurrent ductal carcinoma in situ or invasive cancer.2 This rate is similar to the recurrence rate of invasive cancer treated by wide excision and implies that pathologists are not overdiagnosing ductal carcinoma in situ. It also rebuts the view that a significant number of ductal carcinomas in situ detected by screening would never cause women problems in their lifetime.3

          Thomas considers that the science behind sentinel node biopsy in breast cancer is weak, whereas Beechey-Newman and colleagues feel that the evidence for its use in clinical practice is so compelling that clinical trials are neither necessary nor useful. The latest study, which used technetium labelled sulphur colloid, identified a sentinel node in only 93% of patients, and in those in whom a sentinel node was found the sensitivity was 89%.4 The study's authors admit that the procedure can be technically challenging and that the results varied according to the skill and experience of the surgeon. Although encouraging, these results suggest the technique requires further refinement.

          As Thomas points out, sentinel node biopsy is time consuming and expensive compared with other axillary staging procedures which have been shown to accurately stage the axilla and have little postoperative morbidity.5 The UK study will include an economic evaluation and a comparison of morbidity with sentinel node biopsy and other axillary surgical procedures. Only when the results of this and other studies are available will it be possible to determine whether it is appropriate to introduce sentinel node biopsy into routine practice.

          References

          View Abstract

          Sign in

          Log in through your institution

          Free trial

          Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
          Sign up for a free trial

          Subscribe