Prognostic value of DNA cytometry in 281 premenopausal patients with lymphnode negative breast carcinoma randomized in a control trial - Multivariate analysis with Ki-67 index, mitotic count, and microvessel density

BACKGROUND. The clinical relevance of DNA image cytometry (ICM) and flow cy tometry (FCM) remains under investigation in breast carcinoma. The objectiv e of the current work was to study the prognostic value of DNA ICM and FCM in a series of patients randomized in a control trial. A multivariate analy sis has been performed including other factors still under investigation su ch as Ki-67 index, mitotic count, microvessel density, and P53 and Bcl-2 ex pression. METHODS. Two hundred and eighty-one patients were randomized in the Europea n Organization for Research and Treatment of Cancer 10854 trial comparing s urgery followed by one course of perioperative chemotherapy versus surgery alone. Tumor parameters studied were pT, multicentricity, tumor grading acc ording to modified Scarff-Bloom-Richardson, estrogen receptors, mitotic cou nt per 1.7 mm(2), MIB-1, and BCL-2 scores, microvessel density, and p53 exp ression. ICM DNA parameters studied from paraffin embedded specimens, were DNA ploidy, proliferative index, 2c deviation index, malignancy grade, and Auer-Baldetorp typing. FCM DNA parameters analyzed on the same samples were ploidy and S-phase fraction statistics. The influence of tumor parameters, and DNA parameters on overall survival (OS), disease free survival (DFS), and metastasis-free survival (MFS) was evaluated using the Cox model. Media n follow-up was 82 months. RESULTS. For OS, the prognostic parameters retained were pathologic tumor s ize (pT) and mitotic index (MI). Overall survival was 94% and 68% for tumor s pT1/MI less than 10 and pT2-3 MI greater than or equal to 10, respectivel y. For DFS, age, multicentricity, and grading according to modified Scarff and Bloom were predicting factors with the same relative risk. Disease free survival was 96%, 78% and 68% respectively, when 1, 2, or 3 of those facto rs were present. For MFS, the only retained predicting factor was MI. MFS w as 97% and 73% when MI was less than 10 and MI was greater than or equal to 10, respectively. CONCLUSIONS. Evaluation of proliferative compartment was the most important predicting factor for OS and MFS in the current series of premenopausal ly mph node negative patients with breast invasive carcinoma. When working on paraffin embedded tissue, the best way of assessing it was MI count. ICM DN A analysis results were not selected in multivariate analysis. DNA analysis by FCM should be considered as an unsuitable technique when working on par affin embedded tissue. Cancer 2000;89:174867. (C) 2000 American Cancer Soci ety.