Biological behavior of human breast cancer micrometastases

Purpose: Clinically undetectable micrometastases may account for disease re currence in breast cancer patients after variable disease-free intervals. H owever, little is known about the cellular mechanisms controlling human bre ast cancer micrometastases. We compared tumor proliferation rate, apoptotic index, and angiogenesis in human breast cancer micrometastases with those of macroscopic axillary lymph node metastases. Experimental Design: Seven breast cancer micrometastases (< 2 mm) obtained from the sentinel nodes of seven patients were compared with 13 macrometast ases (lymph node replaced with tumor) obtained from 13 patients. The tissue was fixed in formalin, embedded in paraffin, serially sectioned, and evalu ated by H&E and immunohistochemistry for cytokeratin. Tumor proliferation r ate was assessed as the number of Ki-67-positive nuclei/total number of tum or nuclei. Tumor vascularity was quantified using antibody to factor VIII t o identify microvessels per high-power field (at X400). Apoptosis was quant ified using the terminal deoxynucleotidyl transferase (Tdt)-mediated nick e nd labeling method. Results were analyzed with the Wilcoxon rank-sum test. Results: Median size of micrometastases was 0.5 min (range, 0.4-1.0), and t he median number of tumor nuclei/section was 143 (range, 90-312). Median pr oliferation rate for macrometastases was greater than for micrometastases ( 35% versus 12%; P = 0.003). Median microvessel density/high-power field for macrometastases was greater than for micrometastases (17 versus 1; P < 0.0 01). There was no difference in apoptotic index between macrometastases and micrometastases (1.1% versus 0.7%; P = not significant). Conclusions: Human breast cancer micrometastases have lower tumor prolifera tion rates and angiogenesis than breast cancer macrometastases. These chara cteristics may explain their differential growth patterns.