Automated image analysis of proliferating cells in carcinoma of the larynx

Abstract

Objective—To determine the role of cell proliferation neoplastic progression in the larynx and possibly derive criteria of clinical significance using automated quantitative image analysis.Material and Methods—In a retrospective study involving archival material, the occurrence and location, size, shape and staining intensity of proliferating cell nuclear antigen (PCNA)-positive cells (12,538 cells in total) were analyzed in squamous cell carcinoma (SCC), as well as in pre- and non-neoplastic conditions, using computer-assisted morphometry with reproducibility and sensitivity exceeding 99%.Results—Immunohistochemically detectable PCNA-positive cells were located in the basal layer in non-neoplastic states, in well-differentiated SCCs in layers adjacent to the basal membrane and in poorly differentiated neoplasms in the neoplastic epithelial islets. An increased degree of dysplasia was associated with an increased number of PCNA-immunoreactive cells of increased nuclear size and staining intensity. There was a significant difference between carcinomas and dysplasia in terms of altered nuclear shape. With increasing malignancy of SCCs, nuclear shape alterations and PCNA staining intensity increased, whilst nuclear size decreased.Conclusions—Automated image analysis of cell populations allowed the identification of populations of malignant cells and provided information on the severity of preneoplastic and neoplastic conditions of use in studies of tumor behavior and with potential clinical application.