IMPROVEMENT OF QUALITY-CONTROL FOR STEROID-RECEPTOR MEASUREMENTS - ANALYSIS OF DISTRIBUTIONS IN MORE THAN 40000 PRIMARY BREAST CANCERS

All French laboratories that routinely assay estradiol (ER) and proges terone (PR) receptors participate in the European EORTC quality contro l program based on twice-yearly analysis of 5 cytosolic preparations. This system has considerably reduced inter-laboratory variations, but does not cover all aspects of these assays. Analysis of receptor value distributions is also crucial to ensure that receptor measurements re main stable with time, independently of the laboratory and assay metho d. This study involved 83907 receptor assays carried out in the last 1 7 years by 17 laboratories belonging to the French Study Group on Tiss ue and Molecular Biopathology. The assays were based on radioligand bi nding (RLA) or immunoenzymology (EIA). For each laboratory, the median s and positivity rates were analysed according to two totally objectiv e criteria, the patient's age and the year of assay, and according to histological grade and histological type of the tumor in order to veri fy the correlations classically described. Age-related distributions v aried little between laboratories, compared with data published by 7 E uropean EORTC laboratories [1]. The results remained relatively stable with time in the RLA method for ER and PR, and in the EIA method for PR. Median ER-EIA data showed a marked increase between 1987 and 1989, mainly due to changes in the quality of Abbott reagents during this p eriod. Otherwise, this analysis confirms previous pathophysiological o bservations.