Quantitative morphometric analysis of fine needle aspirates of breast carcinoma.

Background: Reports of morphometric analysis using computerized digitized images of fine needle aspirates of breast cancer are rare. We aimed to evaluate whether quantitative morphometric estimations of cytological nuclear features could be used for preoperative prediction of the risk of recurrence as well as chemosensitivity.
Methods: Seventy consecutive breast cancer patients were treated with preoperative chemotherapy (Doxifluridine 800 mg/body/day, orally) for 4 weeks and subsequently they underwent surgery. Using Papanicolaou-stained cytological materials, computerized morphometric analyses were performed. Mean nuclear area (NA) was measured and the coefficient of variation of NA (NACV) was calculated as the quantitative parameter of nuclear atypism. NA and NACV were compared with prognostic factors (tumor size, histological grade, hormone receptor status, nodal status, and ploidy status), and with the response to the chemotherapy.
Results: NA and NACV were significantly associated with hormone receptor status and ploidy status (p<0.05). NACV correlated with histological grade (p<0.05). Neither NA nor NACV were associated with tumor size and nodal status. Patients with high NACV (>35%) had lower rates of disease-free survival (p<0.05) than those with low NACV (< or =35%). Responders to preoperative chemotherapy had statistically larger NA, higher NACV and higher S-phase fraction at the time of diagnosis compared with non-responders (p<0.001, p<0.0005, and p<0.05 respectively).
Conclusions: Morphometric analysis of preoperative fine needle aspirates reflects important clinical information, such as the risk of recurrence, and particularly, chemosensitivity.