Objective: The goal of this study was to characterize adenocarcinoma in situ (ACIS) and invasive adenocarcinoma (AdCa) of the cervix by using image histometric measurements of nuclear morphometric features.
Study design: Archival pathology slides and tissue blocks from 37 patients with ACIS, 18 with invasive AdCa, and 13 with normal cervical epithelial and glandular histology were reviewed by two pathologists. The controls were matched for age and menstrual status and as closely as possible for the age of the slides; this limited the number of normal cases available. Morphometric, photometric, and textural measurements were made on 4-microm sections of tissue stained with a thionin-SO(2) Feulgen reaction. A mixed analysis of covariance model was used for analysis.
Results: The Integrated Optical Density Index was found between the mean value for normal cells and that for ACIS and invasive AdCa (P <.001). Twenty-two other morphometric features were identified that exhibited differences in their means between at least two of the three tissue types.
Conclusion: In the cell populations studied, certain nuclear image features were found to correlate with histologic diagnosis. The features can be measured objectively and could be useful to pathologists in differentiating lesions, although a larger study should be evaluated to confirm these findings. Further, these features may be important as optical technologies are developed that make diagnoses in real time.