SU-E-T-321: Dynamic Modulated Brachytherapy

We introduced the concept of Dynamic Modulated Brachytherapy (DMBT) for rectal cancer, last year. To continue our work, we studied different shield designs and investigated the system's tolerance against systematic setup errors.

As previously presented, our system uses a cylindrical tungsten shield to create a directional radiation profile, which is modulated through translation and rotation using a specialized robotic arm. We used Monte Carlo simulations and an in-house gradient projection optimization algorithm to look at key design parameters. First, we used ideal phantoms to study treatment quality from shield radii ranging 0.5-1.5 cm in 0.25 cm increments. Second, using 36 patient plans, the dependence on radial source position within the shield was studied. We also analyzed the tolerance of the system to systematic setup errors by simulating dose distributions from possible inaccuracies. These included translational and rotational errors as well as possible Ir-192 source misplacements by the afterloading system.

Changes in shield radius followed steady patterns. Increasing the radius showed a consistent increase in dose conformality to the tumor volume and better sparing to surrounding tissues. However, there was also a linear increase in total dwell time. There was a trade off to changing the radial position of the source. As the source is brought away from the center, there is a decrease in conformality to the tumor volume, but sparing to healthy tissues was increased and there is a decrease in total dwell time. We found that any potential setup errors for our system, within anticipated margins, had negligible effects on the dose distributions (< 3% deviation).

Various parameters for shield designs must be balanced for an effective DMBT application. It was found that the system is highly robust against systematic setup uncertainties.