Background: The incidence of estrogen receptor (ER)-positive breast cancer apparently is increasing. It remains unclear whether this increase is due to an improvement in receptor assay sensitivity, a change in patient characteristics, or a change in tumor biology.
Methods: The distribution of ER, tumor size, and patient age for 11,195 tumor specimens gathered from patients nationwide from 1973 to 1992 were analyzed. All assays were performed in a single laboratory. A single-label, dextran-coated charcoal (DCC) method was used from 1973 to 1984, and a dual-label, DCC method, which allows the determination of both ER and progesterone receptor levels in the same assay, was used from 1985 to 1992.
Results: The median level of ER has increased steadily from 14 fmol per milligram of protein in 1973 to 58 fmol per milligram of protein in 1992 (P < 0.0001). The percentage of ER-positive tumors also rose from 73-78% during the same period (P = 0.008). When the assay method was modified from single to dual label, no abrupt or stepwise increase occurred. Tumor size decreased over the same period (P < 0.0001). From 1973 to 1977, 48% of tumors were larger than 2 cm, and 15% were larger than 5 cm, compared to 60% and 9%, respectively, from 1988 to 1992. The percentage of women older than 50 years of age remained relatively constant over time. After adjusting for tumor size, age, number of positive lymph nodes, and change in assay method, a sustained rise in ER level remained. In a multivariate analysis that included age, age group, year of biopsy, tumor size, and number of positive nodes, the year of biopsy still was independently predictive of ER level (P < 0.0001).
Conclusion: The measured level of ER in primary breast cancers has increased during the last 2 decades. It is unlikely that technical improvements or changes in tumor size, age, or nodal status fully explain this increase. The rising level of ER may reflect a change in breast cancer biology and in hormonal events that influence breast cancer genesis and growth.