Dosimetric analysis of ruthenium-106 ophthalmic applicators for the treatment of retinoblastoma and ocular melanoma

Treatment of retinoblastoma dates back to the first reported (and unsuccessful) enucleation in 1767. Over 200 years later, enucleation is a far more humane and common procedure, yet brachytherapy is often the treatment of choice for small to medium-size tumors offering both function-preservation and good survival-rates (usually employing Co⁶⁰, Ir¹⁹² and I¹²⁵). Use of Ru¹⁰⁶/Rh¹⁰⁶(T_½=1 y; a˜Emax=3.54 MeV) for ocular applicators offers the dosimetric advantage of more rapid dose fall-off and reduced risk of cataracts and optic nerve injury. Ru¹⁰⁶/Rh¹⁰⁶ plaques are therefore suitable for treatment of small tumors, especially in pediatric patients where radiation sensitivity of retinoblastoma is well documented. The authors present dose measurements and Monte-Carlo simulation of the dose distribution for a 14 mm applicator, and compare them with I¹²⁵ plaques. Depth dose and transverse profile distributions were calculated via numerical integration. Beta dose as a function of radial distance from a point source was obtained from the beta dose kernels of Simpkin and Mackie, which are applicable to several beta isotopes and energies. The authors assumed the source was uniformly distributed over the applicator-surface, which was divided into approximately 10⁶ segments. Numerical integration of the kernel was performed over a 1 mm dose-grid. Measurements were performed using both Scanditronix `stereotactic-field diode´ in water phantom, and radiochromic film in water equivalent phantom. Manufacturer specifications are based on measurements with a relatively large detector, and thus have a ±30% uncertainty. In contrast, both measurements as well as the Monte Carlo simulation are in good agreement, and clinically usable