Dosimetric verification of a dedicated 3D treatment planning system for episcleral plaque therapy

Episcleral plaque therapy (EPT) is applied in the management of some malignant ocular tumors. A customized configuration of typically 4 to 20 radioactive seeds is fixed in a gold plaque, and the plaque is sutured to the scleral surface corresponding to the basis of the intraocular tumor, allowing for a localized radiation dose delivery to the tumor. Minimum target doses as high as 100 Gy are directed at malignant tumor sites close to critical normal tissues (e.g., optic disc and macula). Precise dosimetry is therefore fundamental for judging both the risk for normal tissue toxicity and tumor dose prescription. This paper describes the dosimetric verification of a commercially available dedicated treatment planning system (TPS) for EPT when realistic multiple-seed configurations are applied.

The TPS Bebig Plaque Simulator is used to plan EPT at our institution. Relative dose distributions in a water phantom, including central axis depth dose and off-axis dose profiles for three different plaques, the University of Southern California (USC) #9 and the Collaborative Ocular Melanoma Study (COMS) 12-mm and 20-mm plaques, were measured with a diode detector. Each plaque was arranged with realistic multiple 125I seed configurations. The measured dose distributions were compared to the corresponding dose profiles calculated with the TPS. All measurements were corrected for the angular sensitivity variation of the diode.

Single-seed dose distributions measured with our dosimetry setup agreed with previously published data within 3%. For the three multiple-seed plaque configurations, the measured and calculated dose distributions were in good agreement. For the central axis depth doses, the agreement was within 4%, whereas deviations up to 11% were observed in single points far off-axis.

The Bebig Plaque Simulator is a reliable TPS for calculating relative dose distributions around realistic multiple 125I seed configurations in EPT.