Milwaukee's water woes.

The present edition of the British Journal of Cancer contains an article by Stanton et al. on the prognostic significance of DNA ploidy and S-phase analysis in breast cancer. Insofar as hardly a week goes by without a similar manuscript appearing for review, the decision to publish this paper could be questioned, especially as it reports largely negative findings. In this Editorial I would like to discuss the Stanton et al. paper whilst at the same time taking the opportunity to emphasise the need for prognostic factors in breast cancer and to highlight the problems associated with the assessment of indices such as S-phase fraction. The demand for prognostic factors emanates from the great heterogeneity in the natural history of breast cancer and the wide variety of therapeutic approaches used to treat the malignancy. Most breast cancer patients present with evidently local disease. Despite this, the outlook for individual patients varies enormously. Although presenting without evidence of distant disseminated cancer, a proportion of women will die relatively rapidly of metastatic disease; these women presumably had occult lesions at the time of presentation. Conversely many women will survive for 20-25 years even without adjuvant therapy. The spectrum of primary treatment on offer to patients presenting with breast cancer is equally varied, ranging from local removal of tumour to high dose chemotherapy with bone marrow rescue. There is a need to match individual patients with appropriate treatments to avoid the equally unacceptable scenaria of either under-treating patients who have inherently aggressive disease or exposing others with indolent tumours to the unnecessary toxic side-effects of potentially ineffective treatment. Despite the need for predictors of prognosis, the only widely utilised factors are related to clinical staging. In particular , the histological presence or absence of metastatic deposits of tumour in axillary lymph nodes markedly affects outcome. The poorer prognosis of patients with invaded lymph nodes has led to official recommendation that most of these patients should receive some form of adjuvant treatment as part of their primary management (Clinical Alert from the National Cancer Institute, May 16th 1988). Conversely , since approximately 70% of patients without lymph node involvement survive long-term, it has been argued that this group could be spared aggressive therapy. However, there are substantial minorities whose disease behaves exceptionally, and it has been suggested that lymph node status largely monitors the extent of disease. Hence the need for additional factors which …