Abstract

Aims—The use of the H score (involving the assessment of intensity and distribution of positivity) on sections stained for the oestrogen receptor (ER) by immunocytochemistry (ICC) allows different samples to be compared and detailed correlations to be made between hormone receptor expression and morphology. This study assessed the reliability of core biopsy in predicting ER expression in the same tumour excised later at treatment. The distribution of ER within excised tumours was investigated.

Methods—The distribution of ER positivity was investigated in 51 diagnostic core biopsies and across the diameter of 51 subsequently excised tumours in a field by field (magnification, ×40; field diameter, 0.4 mm) assessment using the semiquantitive H scoring system.

Results—The ER H score in diagnostic core biopsy was significantly higher (p = 0.05, paired rank test; overall mean, 130; n = 51) than the mean in the corresponding excised tumour (mean, 110; n = 51). There was a significant downward trend in ER positivity from the periphery of tumours towards the centre (p = 0.001). The reduction of ER positivity was 6 H score units (2%)/mm. If core biopsies were orientated with the tumour edge at one end no change in ER positivity with field number along the length of the core could be demonstrated.

Conclusions—ER estimation in core biopsies correlated well with expression in tumours but ER expression was higher in the core biopsies than in the excised tumours. ER expression was higher at the periphery of tumours than at the centre. The higher ER expression in cores may reflect the higher chance of sampling the peripheral part of a tumour using a needle core.