An immunocytochemical assay using monoclonal anti-progesterone receptor (PR-ICA) was performed in nonmalignant (N = 57) and malignant (N = 200) breast disorders. The results were analyzed with a computerized system of image analysis referred to as SAMBA and correlated with binding assays (DCC), and with standard histopathologic findings. It was shown that there was a correlation between 1) the PR-ICA and the binding assays (91.5%), and 2) between the binding assays and the multiparametric computerized (SAMBA) analysis of the PR-ICA. It was also shown that SAMBA provides an accurate, reliable, and reproducible evaluation of PR-ICA that is complementary to binding assays and constitutes a standardized method of evaluating the heterogeneity of the progesterone receptor (PR) distribution in tumors. It is concluded that SAMBA analysis of both the PR-ICA and estrogen receptor (ER-ICA) should improve the prognostic evaluation and the prediction of responsiveness to endocrine therapy in breast carcinomas.