The relationship between hormone receptor status and the effect of adjuvant tamoxifen in early breast cancer remains controversial. This article presents the results of a randomized trial of adjuvant tamoxifen (40 mg daily for 2 years) versus no adjuvant endocrine therapy in postmenopausal patients. During 1976 to 1984, 1,407 patients were included in the study. Of these, 427 (30%) had high-risk tumors (pN + or pT greater than 30 mm) and were included in a concurrent randomized comparison of postoperative radiotherapy versus adjuvant polychemotherapy. The mean follow-up time was 61/2 years. Tamoxifen improved the recurrence-free survival (RFS) (P less than .01), but the overall survival difference in favor of the tamoxifen-allocated patients was not significant. Data on estrogen (ER) and progesterone receptor (PgR) content were available in 750 patients. Their mean follow-up time was 41/2 years. The effect of tamoxifen was significantly related to ER level (P less than .01). No benefit with tamoxifen was observed among ER-negative patients. The relation to PgR level was of borderline significance (P = .06). Multivariate analysis indicated that most of the interaction between treatment and receptor content was explained by the interaction with ER (P less than .01). The PgR status appeared to modify the effect of tamoxifen among the ER-positive patients and the greatest effect was observed among patients who were positive for both receptors. However, the additional predictive information provided by the PgR assay did not help to identify an unresponsive subgroup of patients.