GRADING AND PROGNOSIS OF INFILTRATING DUCTAL BREAST-CARCINOMA BY MEANNUCLEAR VOLUME ESTIMATES

OBJECTIVE: Quantitative evaluation of nuclear size in breast carcer is performed with the aim of investigating whether the data obtained are related to subjective grading and prognosis. STUDY DESIGN: Unbiased s tereologic estimates of the volume-weighted mean nuclear volume (nucle ar (v) over bar(v)) in paraffin-embedded tissue from 44 specimens of i nfiltrating ductal breast carcinoma were obtained by the point-sampled intercepts method. RESULTS: Histologically graded cases showed an ove rlap in nuclear (v) over bar(v) estimates, but significant differences were demonstrated in which averaged values increased from grade 1 to 3 breast carcinoma (mean, 495.4 mu m(3); SD, 101.7 mu m(3)) were signi ficantly larger than those in grades 1 (mean, 327.8 mu m(3); SD, 90.9 mu m(3)) and 2 (mean, 409.9 mu m(3); SD, 72.6 mu m(3)) (P < .01). Mino r statistically significant differences (P < .05) were found between g rades 1 and 2. Estimates of nuclear (v) over bar(v) based on systemati c sampling within the specimen showed high efficiency, with >80% of th e total observed variance contributing to biologic differences between individual specimens. Univariate prognostic study showed that clinica l stage, and not histologic grade, was of prognostic value in the case s studied. Similarly, nuclear (v) over bar(v) estimates were associate d with disease recurrence and survival, with a cutoff point of 425 mu m(3). CONCLUSION: Based on the cases investigated, shape-independent n uclear (v) over bar(v) estimates may be an adjuvant tool in the gradin g and prognostic evaluation of infiltrating ductal breast carcinoma.