Search

Purpose: To determine and compare cyclo-oxygenase-2 (COX-2) expression in photon-radiated and non-radiated malignant uveal melanomas and to analyse the correlation between COX-2 expression and prognosis. Methods: Immunohistochemical staining for COX-2 was performed on 21 uveal melanomas that were endoresected after prior stereotactic radiotherapy with photons and on 22 tumours that were treated by endoresection without prior radiotherapy. COX-2 staining was further analysed in respect to cell type, maximal prominence, time interval between radiotherapy and surgery, apoptotic index (AI), proliferative index (PI) and the development of metastatic disease. Results: There was no difference in COX-2 expression between radiated and non-radiated melanomas (P > 0.15). COX-2 staining correlated with neither the tumour prominence (P > 0.40) nor the AI or the PI (both P > 0.35). Tumours with high COX-2 expression were significantly more likely to develop metastasis (P = 0.022). Conclusion: Radiotherapy with photons does not induce COX-2 expression in malignant melanomas of the uvea. But high COX-2 expression may be a marker for poor prognosis.

A modified swim goggle holding a light spot as an optical guide for actively aligning the eye in a reproducible orientation has been constructed to perform radiotherapy of ocular tumors. This device is compatible with computed tomography (CT) and magnetic resonance imaging systems. Image fusion of these data sets yielded clinically acceptable results. The reproducibility of the eye's positioning is tested by repeated CT. The eye's alignment during radiotherapy is monitored by an infrared TV camera with individual markings of the eye's position on the TV-monitor screen. From 2003-2006, 50 patients were treated with this fixation aid by radiosurgery with good patient compliance.