Immunohistochemically determined
hormone receptors have gained importance as prognosticators in primary
breast cancer, but their definitive role has not yet been evaluated.
METHODS:
Tumor samples from 299 patients were examined for
estrogen and
progesterone receptors by biochemical and immunohistochemical assay. Correlations with established risk factors (
tumor size, lymph node status, menopausal status, grading including subfactors) and histopathology were analyzed.
RESULTS: The
estrogen receptor, determined by immunohistochemical method revealed positivity in 80.6% of patients; biochemical measurement yielded 76.2% positive results. The
progesterone receptor measured by immunohistochemistry yielded 61.3% positivity versus 55.8% detected by biochemical analysis. Invasive lobular, tubular, and ductal invasive
carcinoma with prominent stroma content ("
scirrhous carcinoma") rather than ductal invasive
carcinoma was more frequently
estrogen-receptor positive with immunohistochemistry than with biochemical assay. For
progesterone receptor, the same pattern of positivity was seen with immunohistochemical assay. With
progesterone receptor determined biochemically, "scirrhous" and
lobular carcinoma showed positive results in a lower proportion than invasive ductal and
tubular carcinoma. Significant correlations were observed between the
estrogen-receptor status, the histologic grade of
malignancy, nuclear polymorphism, and the rate of mitosis with both methods (p < 0.001 respectively). Different correlations were found between
tumor size, menopausal status and
estrogen receptor status with both assays respectively. For the
progesterone receptor status, immunohistochemistry yielded significant correlations with the histologic grade of
malignancy, nuclear polymorphism, rate of mitosis (p < 0.001 respectively) as well as growth pattern (p < 0.01), while biochemical analysis revealed a correlation with nuclear polymorphism (p < 0.05). The correlation analysis of both components of the immunoreactive score revealed a more significant impact of percentage of positive cells than of staining intensity.
CONCLUSIONS: Immunohistochemistry detected a closer correlation between prognostic factors and receptor data than biochemical analysis.