Clinical Update
Breast cancer and distant metastatic disease

CMAJ 2000;163(2):203-4

Background

Many women with breast cancer that appears localized to the breast or to the breast and regional lymph nodes are believed to have undetectable distant metastases at the time of initial staging. Because of this, current recommendations are that systemic chemotherapy be given to all women with regional node involvement and to those with no node involvement but a primary lesion greater than 1 or 2 cm in diameter. However, the improvement in outcome with chemotherapy is slight. In women with a primary tumour of 1 cm or less, systemic therapy reduces the rate of distant metastases from 10% to 7%. Although this is a risk reduction of 40%, it means, in effect, that an average of only 3 women on average in every 100 actually benefit from the therapy.¹ Thus, many women may needlessly undergo chemotherapy. However, if distant metastatic disease could be detected at the time of initial cancer diagnosis, the precision of prognistic stratification might be improved. A new technique to detect bone marrow micrometastatic lesions is now available.

Question

What proportion of women presenting with stage I, II or III breast cancer have distant metastatic disease?

Design

Bone marrow aspirates from the upper iliac crests were obtained from 743 consecutive patients admitted to a breast clinic in Germany; 552 had been newly diagnosed with stage I, II or III breast cancer and 191 patients with nonmalignant disease (mainly of the breast). The patients were followed for several years, thus providing some information on the test's usefulness in categorizing women into meaningful risk groups.

Results

Using an immunocytochemical technique to detect an antigen on cytokeratin peptides (a specific marker of epithelial cancer cells in bone marrow), 36% of the women with breast cancer tested positive, compared with only 1% of women without breast cancer. During a median follow-up period of 38 months, women with marrow-detectable cytokeratins were much more likely to die of cancer-related causes than women without the markers (relative risk 4.17; 95% confidence interval 2.51—6.94; p < 0.001). There were 301 women with no detectable lymph-node metastastes. Cytokeratin markers were found in the bone marrow of 23% of women with tumours smaller than 0.5 cm and 35% of those with tumours 0.5 to 1.0 cm. Of the 100 patients with node-negative tumours and evidence of micrometastases, 14 died of cancer-related causes over the follow-up period, compared with only 2 of the 201 women without micrometastases.

Commentary

This is a carefully conducted study of a large cohort of women. The testing for cytokeratins was done independently from the clinical staging. However, follow-up was relatively short, especially for a disease such as breast cancer.

Clinical implications

In a related editorial,¹ Barbara Smith comments that it is now important to proceed with trials of therapy because the follow-up is relatively short for breast cancer. Nonetheless, the survival curves extending to 48 months show a continuous divergence of survival and disease-free survival between groups with and without micrometastases. Smith feels that it is premature to recommend cytokeratin tests to patients with disease limited to the breast who wish to avoid chemotherapy. — John Hoey, CMAJ

Send a letter to the editor Envoyez une lettre à la rédaction

1. Smith BL. Approaches to breast-cancer screening. N Engl J Med 2000;342(8):580-1.

© 2000 Canadian Medical Association or its licensors