Cytological evaluation of biological prognostic markers from primary breast carcinomas

This study was undertaken to evaluate our ability to detect multiple molecular markers of prognosis and response to treatment in fine needle aspirates (FNA) from patients with primary breast carcinomas. 147 patients with operable primary breast carcinomas who had been recruited to a randomized trial of primary medical therapy (PMT) versus adjuvant chemoendocrine therapy were analysed. FNAs were taken prior to therapy and from this multiple slides were produced using cytospin cytocentrifugation and stored at -80 degrees C for subsequent immunocytochemical analysis (ICA). ICA was performed for oestrogen receptor (ER), progesterone receptor (PgR), p53, Ki67, and Bcl-2. Part of the aspirate was snap frozen and used for flow cytometric analysis of ploidy and S-phase fraction (SPF). In a subgroup of 50 patients who had surgery prior to systemic therapy, as well as FNAs, sections were also taken from paraffin-embedded blocks and stained by ICA for ER, PgR and p53 for validation. In these patients ER was additionally measured by enzyme immunoassay (EIA) from frozen tissue taken at surgery. ER, PgR, p53, Bcl-2, and Ki67 were successfully detected by ICA while ploidy and SPF were successfully measured by flow cytometry from FNA material. The percentage positive values obtained were reasonable and as follows: 74% for ER, 70% for PgR, 36% for p53, 80% for Bcl-2,68% of tumours were aneuploid and 32% diploid. Significant relationships between these measurements were observed in accordance with expectations. The concordance for ER, PgR, and p53 from FNA when compared to ICA of matching histological sections was 91.5%, 75.5%, and 75% respectively. For ER the concordance between measurement by ICA of cytological and histological samples and by EIA of frozen tissue was 82.5% and 84% respectively. These results indicate that multiple molecular markers can be adequately tested on cytological preparations from primary breast tumours. These markers can be used to determine prognosis and predict response to PMT.