Prognostic significance of estrogen receptors in 405 primary breast cancers: a comparison of immunohistochemical and biochemical methods

Over the last few years, estrogen receptor determinationby means of immunohistochemistry has been extensively used.The aim of this study was to comparethis technique with estrogen receptor determination by meansof dextran-coated charcoal, and to evaluate whether oneof the two methods is more predictive ofprognosis. Estrogen receptors were determined by means ofboth the dextran-coated charcoal method and immunohistochemistry in405 patients with primary breast cancer; age, pathologicaltumor size, nodal status, and progesteron receptors bydextran-coated charcoal method were also recorded. The disease-freeand overall survival probabilities were estimated using theproduct-limit method; Cox's proportional hazard model was usedto evaluate the prognostic role of estrogen receptorsas determined by the two methods.There appears to be a close association betweenestrogen receptor determination by the two methods (81.5%of concordant results) and their prognostic role wassimilar, even when the patients were divided intodifferent groups (on the basis of their estrogenreceptor status) and adjustments for the effect ofother prognostic variables were taken into account. Ourstudy shows that the two methods can beused indifferently to evaluate estrogen receptor status asa prognostic factor in breast cancer patients.