Search

This study investigated two hypotheses: (1) sufficient cells may be obtained by needle aspiration of breast nodules to produce good flow cytometric DNA profiles; and (2) benign breast lesions do not produce aneuploid G0G1 peaks, and therefore a distinct aneuploid peak is sufficient for a diagnosis of malignancy. Breast specimens received in Surgical Pathology between December 1985 and February 1987 were aspirated, and the cells stained with propidium iodide for flow cytometric DNA analysis. A total of 344 specimens were aspirated, of which 204 (59%) were malignant and 140 (41%) benign. One hundred fifty-three malignant and 111 benign specimens contained sufficient cells for analysis. Cytologic smears were available for 177 malignant and 123 benign specimens. DNA histograms were considered diagnostic of malignancy if an aneuploid peak was present which contained at least 20% of the cells in the distribution, and had a DNA index greater than or equal to 1.2. Using these criteria, 73 of 153 (48%) carcinomas could be identified. None of the benign lesions satisfied these criteria. One fibroadenoma with atypical hyperplasia produced a distinct peak which contained less than 5% of the cells in the histogram, and had a DNA index of 1.25. Flow cytometric analysis provides objective data that complement the subjective cytologic interpretation of fine needle aspirates.

Flow cytometric DNA analysis was performed on 168 consecutive primary operable breast cancers and the results correlated with thymidine labeling, estrogen receptor (ER) and progesterone receptor values, and various histologic parameters. Seventy-five cancers (45.0%) were diploid and 93 (55.0%) aneuploid. In 83.9% of aneuploid cancers, the DNA index fell between 1.1 and 2.0. Thirteen cancers were hypertetraploid and three hypodiploid. The percent of S-phase cancers (SpF) varied from 1.1% to 24.4%, with a mean of 8.0% and a median of 7.1%. The mean SpF of diploid cancers was 5.2%; of aneuploid cancers, 10.3%. There was no significant correlation between SpF or ploidy and tumor size or axillary lymph node status. The thymidine labeling index (TLI) varied from 0.2 to 23.1, with a mean of 7.5 and a median of 6.1. There was good correlation between TLI and SpF (r = 0.892, P = 0.0001). ER-negative tumors had a significantly higher mean SpF (10.3%) than did ER-positive tumors (6.7%), but there was no significant correlation between ploidy and receptor positivity or negativity. There was a good correlation between invasive tumor necrosis, poor cytologic differentiation, aneuploidy, and above-median SpF. Only a fair correlation was observed between mitotic rate and SpF. A classification of invasive breast cancers based on ploidy and SpF is proposed.