Program Information
Dosimetric Characteristics for a Low-Dose-Rate Brachytherapy Stent for Esophageal Cancer
Mark J Rivard1*, Arnold M Herskovic2, John H Hingston3, Claude Clerc3, John T Favreau3, (1) Tufts University, Boston, MA, (2) Rush University Medical Center, Chicago, IL, (3) Boston Scientific, Marlborough, MA
Presentations
TU-C3-GePD-T-2 (Tuesday, August 1, 2017) 10:30 AM - 11:00 AM Room: Therapy ePoster Lounge
Purpose: Esophageal cancer afflicts 0.5M people annually with half the patients inoperable at detection. Radiation therapy generally takes several weeks and irradiates healthy tissues with large treatment margins, yet 5-year survival rates are <20%. Esophageal stents provide immediate dysphasia relief and can be used alone or with EBRT. Combining the benefits of stenting with radiation, an LDR brachytherapy stent was designed based on Monte Carlo dosimetry.
Methods: To match the physical stent geometry, the simulated stent lengths and outer diameters spanned 12cm and 2cm. Plastic tubes attached inside helically-braided stents (plastic or metal) contained stranded LDR brachytherapy seeds (¹²⁵I or ¹³¹Cs). Given the stent and air-filled interior were not radiologically-equivalent to water (or tissue), a brachytherapy TPS would not suffice and Monte Carlo simulations using MCNP6 provided accurate dosimetry. To homogenize surface dose and minimize dose hotspots to the radiation-sensitive mucosa, seed density and source-strength weighting were varied. Doses were calculated 0-2.5cm from the stent and ±4.5cm along the stent length with 0.01cm² voxels and 2° azimuthal resolution. Doses were normalized to the 0.5cm target depth at stent midplane.
Results: Compared to the standard design, surface dose uniformity improved by increasing seed density and source-strength weighting, producing dose hotspots of 377%, 300%, and 361%, respectively. In combination with source-strength weighting, the maximum surface dose decreased to a 278% dose hotspot. In practice, all dose hotspots would be diminished due to breathing and peristalsis. The plastic stent attenuated less than the metal stent and produced >16% lower dose hotspots. ¹³¹Cs seeds produced dose distributions that had better surface and target dose uniformity than ¹²⁵I; however, the radiation was more penetrating and the radiobiological influence of the shorter half-life needs to be evaluated in practice.
Conclusion: The radioactive stent design provides adequately-uniform dose distributions and compatibility with anatomic geometry.
Funding Support, Disclosures, and Conflict of Interest: Research support for this project is provided by Boston Scientific, Inc. (Marlborough, MA).
Contact Email: