QUANTITATIVE MEASURES IN BREAST CANCER PATHOLOGY

我们的目标是建立预测的细胞形态计量标准

Our goal is to establish cytomorphometric criteria for prediction of breast cancer recurrence in Stage I disease. Several studies have suggested the prognostic merits of individual morphometric parameters in breast cancer and occasional reports have illustrated the discriminatory power of these variables in comparison with conventional prognostic factors. However, until now the necessity for special specimen handling, laborious methods, and/or expensive computers for quantitative cell image analysis have limited the application. Our approach is to employ a new form of automated analysis termed "contextual analysis" which permits quantification of architectural features and combines them with conventional computer high resolution single cell analysis. Preliminary studies show that contextual analysis provides information complementary to single cell analysis. The study of contextual features will be a major developmental aspect of this study, since they have yet to be employed for determination of prognosis. Fine needle aspirate smears and corresponding tissue biopsy specimens from more than 150 patients with Stage I disease, followed over 5 years and being analyzed for pertinent histological, clinical and epidemiologic factors, will be employed. Cases will be unifrom with respect to staging and treatment protocols. Specimens will be acquired from two centers and feature analysis will be independently confirmed by both laboratories. Thus, the proposed study offers a novel and economical approach to assess the merits of morphometry in Stage I breast cancer prognosis.