Over the last few years, estrogen receptor determination by means of immunohistochemistry has been extensively used. The aim of this study was to compare this technique with estrogen receptor determination by means of dextran-coated charcoal, and to evaluate whether one of the two methods is more predictive of prognosis. Estrogen receptors were determined by means of both the dextran-coated charcoal method and immunohistochemistry in 405 patients with primary breast cancer; age, pathological tumor size, nodal status, and progesteron receptors by dextran-coated charcoal method were also recorded. The disease-free and overall survival probabilities were estimated using the product-limit method; Cox's proportional hazard model was used to evaluate the prognostic role of estrogen receptors as determined by the two methods. There appears to be a close association between estrogen receptor determination by the two methods (81.5% of concordant results) and their prognostic role was similar, even when the patients were divided into different groups (on the basis of their estrogen receptor status) and adjustments for the effect of other prognostic variables were taken into account. Our study shows that the two methods can be used indifferently to evaluate estrogen receptor status as a prognostic factor in breast cancer patients.