Manual versus image analysis estimation of PCNA in breast carcinoma

To compare manual to image analysis estimation of proliferating cell nuclear antigen (PCNA) expression in paraffin sections of breast carcinomas.

Paraffin sections of 51 breast carcinomas were stained with primary antibody to PCNA. Nuclear PCNA expression in 100 randomly selected tumor cells from marked areas was manually graded from 0 to 3. Antigen expression was also calculated by a cell analysis system (CAS-200, Becton Dickinson, Elmhurst, Illinois, U.S.A.) from marked and random microscopic fields. Obtained proliferative index (PI) from both methods was compared.

Manually calculated PI correlated strongly with the CAS-200-calculated PI (P < .01). The highest correlation was seen between the CAS-200 PI value and manually calculated PI value using grade 2 and 3 nuclei. A particularly high correlation was noted between the number of positive nuclei and antigen staining area (P < .01) as estimated by the CAS-200.

Nuclear expression of PCNA and other nuclear antigens can be accurately evaluated by an image analysis system. The speed and objectivity of such machines allow the evaluation of larger parts of tissues and provide more-representative antigen expression profiles.