Click here to


Are you sure ?

Yes, do it No, cancel

Influence of Convex Curvature for Improved Post-Surgical Dosimetry of a Directional Pd-103 Brachytherapy Implant

Reshma Munbodh1* and Mark J Rivard1, (1) Alpert Medical School of Brown University, Providence, RI


(Tuesday, 7/31/2018) 10:30 AM - 11:00 AM

Room: Exhibit Hall | Forum 3

Purpose: Brachytherapy using permanent implantation with an array of LDR Pd-103 directional sources (CivaSheet) has been used in dozens of patients. However, the influence of implant curvature on dose distributions and volumetric coverage has not been evaluated. This study examined an assortment of implant curvatures towards generalizing the importance of this effect for improved dosimetry.

Methods: A 6x10 source array was examined with 1.5-, 2-, 3-, 4-, 5-, 7-, and 10-cm curvature radii. The 6-source alignment was along a cylindrical long-axis while the 10-source alignment was around a cylindrical circumference with radiation pointing outwards (convexly). Additionally, a planar array (reference geometry) was examined to evaluate implant curvature influence on dose for the unshielded side of the device to a 5-mm depth. Target coverage was assessed with DVHs for 8-, 4-, and 0-mm margins from the array as well as with triangulation on the implant corners to account for dose falloff. Dose ratios of the curved implant relative to the planar implant were evaluated as a function of depth and curvature. Dose was calculated using MCNP6 in a 20-cm radius water phantom and 0.5-mm voxels.

Results: Compared to the planar array, dose distributions and ratios in the vicinity of the implants decreased with smaller curvature radii and greater distances from contact with the CivaSheet. Dose decreased by 29% and 21% between midline and peripheral lines of 6 sources oriented along the 1.5-cm radius cylinder long-axis at a 5-mm depth. These diminishments reduced with increasing array curvature approaching the reference planar geometry. For all array curvatures, target volumetric coverage improved with smaller regions, being nearly constant with minimal margin beyond the source positions.

Conclusion: This study will refine source-ordering nomograms to account for curved implant geometry as approximated preceding surgery for improved clinical target coverage, surely better than a simple planar approach.

Funding Support, Disclosures, and Conflict of Interest: Dr. Rivard has served as a consultant to CivaTech Oncology, manufacturer of the source examined in this study.


Geometric Distortion, Monte Carlo


TH- Brachytherapy: Dose optimization and planning

Contact Email