Abstract
Abstract: :
Purpose: The co-expression of the intermediate filaments vimentin (V) and cytokeratin (Ck), referred to as the interconverted phenotype (IP), confers increased invasive potential to tumor cells and has been shown to be a marker of poor prognosis in uveal melanoma. COX-2, a prostaglandin synthetase, has also been shown to play a major role in processes such as tumor invasion, angiogenesis, and metastasis. The objective of this study is to investigate the correlation between COX-2 expression and IP in uveal melanoma. Methods: One hundred cases of uveal melanoma were sent from the American Registry of Ocular Pathology at the Armed Forces Institute of Pathology (AFIP), Washington, D.C. to the Henry C. Witelson Eye Pathology Laboratory, Montreal, Canada. The formalin-fixed, paraffin-embedded cases were immunostained with monoclonal antibodies against V, Ck and COX-2. A tumor was considered positive for IP when both V and Ck were recorded as positive in tumor cells. Using the chi square test for contingency tables, IP was correlated to COX-2 in malignant cells as well as COX-2 in macrophages. Results: Forty-three out of 100 cases were positive for both V and Ck. Seventy out of 100 cases were COX-2 positive in macrophages. Fifty-eight out of 100 cases were COX-2 positive in tumor cells. Of the 43 cases positive for IP, 27 cases were also COX-2 positive in macrophages. Analysis showed a statistically significant correlation (p=0.00011761) between IP and macrophage COX-2 expression. Of the 43 cases positive for IP, 25 cases were also COX-2 positive in tumor cells. There was a statistically significant correlation (p=0.03388594) between IP and tumor cell COX-2 expression. Conclusion: To the best of our knowledge this is the first study to show a correlation between the interconverted phenotype and COX-2 expression in uveal melanoma. Tumor cells expressing IP and COX-2 have an increased ability to invade the extracellular matrix (ECM) and COX-2 expression of ECM cells such as macrophages may enhance this invasive potential.
Keywords: 350 clinical (human) or epidemiologic studies: biostatistics/epidemiology methodology • 464 melanoma • 496 oncology