PURPOSE: Increased level of cyclooxygenase-2 (COX-2) plays a significant role in the pathogenesis of cancers. High expression of COX-2 has been demonstrated in several cancer types including retinoblastoma. However, the in vivo study did not confirm the anti-proliferative effect of COX-2 inhibitor, celecoxib, on a murine transgenic retinoblastoma model. We, therefore, aim to investigate COX-2 expression in paraffin-embedded retinoblastoma specimens in a larger study group. METHODS: We reviewed 55 retinoblastoma specimens obtained during 1995 to 2005. Clinical and histopathological data were recorded. Immunohistochemical evaluation of COX-2 expression was performed using a rabbit monoclonal antibody to human cyclooxygenase-2. RESULTS: Forty-four of 55 specimens (80%) showed negative immunoreactivity for COX-2 expression. For the 11 specimens (20%, 95% CI = 11.6-32.4%) with positive COX-2, all immunostainings were less than 50% of tumor area. Demographic data and treatment details were available in 53 specimens. Enucleation was performed as a primary treatment in 43 specimens (81%). Other treatments, mainly systemic chemotherapy, were given prior to enucleation in 10 specimens (19%). There was no statistical difference in COX-2 expression between the specimens identified as primary and secondary enucleation (p = 0.66). Regarding the histopathological findings, there were no significant differences between COX-2 negative and COX-2 positive groups. CONCLUSIONS: It appears that COX-2 is not overexpressed in our retinoblastoma specimens, which is different from previous studies. This conflicting data reduces the possibility of introducing Cox-2 inhibitors in the treatment of retinoblastoma.
PURPOSE: Increased level of cyclooxygenase-2 (COX-2) plays a significant role in the pathogenesis of cancers. High expression of COX-2 has been demonstrated in several cancer types including retinoblastoma. However, the in vivo study did not confirm the anti-proliferative effect of COX-2 inhibitor, celecoxib, on a murine transgenic retinoblastoma model. We, therefore, aim to investigate COX-2 expression in paraffin-embedded retinoblastoma specimens in a larger study group. METHODS: We reviewed 55 retinoblastoma specimens obtained during 1995 to 2005. Clinical and histopathological data were recorded. Immunohistochemical evaluation of COX-2 expression was performed using a rabbit monoclonal antibody to humancyclooxygenase-2. RESULTS: Forty-four of 55 specimens (80%) showed negative immunoreactivity for COX-2 expression. For the 11 specimens (20%, 95% CI = 11.6-32.4%) with positive COX-2, all immunostainings were less than 50% of tumor area. Demographic data and treatment details were available in 53 specimens. Enucleation was performed as a primary treatment in 43 specimens (81%). Other treatments, mainly systemic chemotherapy, were given prior to enucleation in 10 specimens (19%). There was no statistical difference in COX-2 expression between the specimens identified as primary and secondary enucleation (p = 0.66). Regarding the histopathological findings, there were no significant differences between COX-2 negative and COX-2 positive groups. CONCLUSIONS: It appears that COX-2 is not overexpressed in our retinoblastoma specimens, which is different from previous studies. This conflicting data reduces the possibility of introducing Cox-2 inhibitors in the treatment of retinoblastoma.