
Determination of proliferation index with MIB-1 in advanced ovarian cancer using quantitative image analysis.
A new monoclonal antibody, MIB-1, reacts with the same epitope recognized by Ki-67. The authors investigated the feasibility of using image analysis to quantitate the MIB-1 staining (proliferation index [PI]) of epithelial ovarian cancers. The PI was determined in 50 advanced-stage primary ovarian cancers. Paraffin sections were immunostained with the MIB-1 monoclonal antibody, and the PI was calculated using a CAS 200 image analyzer. Among 36 stage III ovarian carcinomas, the median PI was 15.1%, compared with 18.9% in 14 stage IV cancers (P = .47). Based on exploratory methods, a cutoff point of 7% best dichotomized these patients into two prognostic groups. Of 39 patients whose cancers had a high MIB-1 expression (> or = 7%), the median survival was 16.5 months, which differed significantly (P = .01) from the median survival of 33.2 months observed in the 11 patients whose tumors demonstrated low MIB-1 expression (< 7%). Using MIB-1 as a binary variable, a strong correlation was found between overexpression of c-erbB-2 (2+ and 3+) and MIB-1 > or = 7% (P = .001). No relationship was found between PI and histologic grade. Further studies are warranted to investigate the relationship between MIB-1, PI expression, and other known clinicopathologic and genetic features of early- and late-stage ovarian cancer.
Duke Scholars
Published In
DOI
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Pathology
- Ovarian Neoplasms
- Nuclear Proteins
- Neoplasm Proteins
- Mitotic Index
- Leukocyte Common Antigens
- Ki-67 Antigen
- Immunohistochemistry
- Image Processing, Computer-Assisted
- Humans
Citation

Published In
DOI
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Pathology
- Ovarian Neoplasms
- Nuclear Proteins
- Neoplasm Proteins
- Mitotic Index
- Leukocyte Common Antigens
- Ki-67 Antigen
- Immunohistochemistry
- Image Processing, Computer-Assisted
- Humans