HDRMC, An Accelerated CT-Based Monte Carlo Dose Calculator for HDR Brachytherapy
O Chibani1,2*, C Ma2, (1) King Faisal Specialist Hospital, Riyadh, ,(2) Fox Chase Cancer Center, Philadelphia, PASU-E-T-519 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To introduce a new CT-based Monte Carlo dose calculation engine (HDRMC) for HDR brachytherapy planning. Dose distributions for gynecologic and esophageal cancer patient plans are calculated and compared to results from a commercial treatment planning system.
Methods:HDRMC is an accelerated Monte Carlo code similar to the MCPI code (Chibani and Williamson, Med. Phys. 32, 3688-3698, 2005). It accounts for tissue and non-tissue heterogeneities. A CT-based applicator model is used. Both Fletcher and Ring gynecologic applicators are considered. HDRMC is benchmarked against the TPS under the assumption of infinite water medium. HDRM is used to quantify the effects of three assumptions commonly used in 3D brachytherapy planning (Infinite geometry, Water-equivalent geometry, and applicator and contrast medium can be part of contoured structures) on DVH parameters.
Results:Dose distributions based on applicator models re-constructed using 1 vs. 2-mm voxels are perceptibly the same for both Fletcher and Ring applicators. Including tissue heterogeneities for the esophageal case decreases CTV D90 by 3.1% and maximum spinal cord dose by 3.6%. Using the infinite geometry assumption in cervical cancer cases can lead to 1.8% and 3% overestimate of D90 and D100, respectively. Using the water medium assumption causes D90, D100, V100, V150 and V200 to be overestimated by 2.7, 4.4, 4.3, 5.6, and 6.2%. D90 value can be overestimated by 4.5% when part of the applicator is included in the CTV. Combining the three effects, can result in 7.4% under-dosage of CTV and 25% overestimate of V100. For the bladder, excluding the contrast medium from the bladder volume can reduce D2cc by 17.5%.
Conclusion:CT-based applicator models can be used for CT-compatible applicators. Dosimetric effects due to non-tissue heterogeneities are not negligible. The inclusion of non-tissue object in contours of CTV and OARs can have a significant effect on DVH parameters.