ESTRO 2024 - Abstract Book

S371

Beachytherapy - Physics

ESTRO 2024

RapidBrachyIVBT, a Monte Carlo dose calculation software based on the Geant4 toolkit v. 10.02.p02 was developed and integrated into RapidBrachyMCTPS, a brachytherapy treatment planning system (12). A Novoste Beta-Cath 3.5F IVBT device with a 90 Sr 90 Y source train was simulated. The software was validated with TG-149 parameters compared to Monte Carlo simulations in water. RapidBrachyIVBT was tested with OCT images from a patient undergoing coronary IVBT for recurrent in-stent restenosis at Brigham and Women’s Hospital in Boston, Massachusetts. The patient artery lumen, calcified plaque, media, stents, and guidewire were segmented. Simulations were performed on the Digital Research Alliance of Canada Cedar computing cluster with 1 billion decay events to yield 0.4% uncertainty on absorbed dose in the target volume for the patient simulation, and 100 million decay events to yield 1% uncertainty at the target volume for the validation simulation. The absorbed dose was scored in rectangular voxels of 0.1 x 0.1 x 0.5 mm 3 for the validation simulation and 0.1 x 0.1 x 1.0 mm 3 for the patient simulation. The prescribed dose was normalized to 23 Gy at 2.0 mm, the target volume, from the source center in water.

Results:

Dose rate in water values using RapidBrachyIVBT agreed with TG-149 consensus dose rate values within 0.07% - 1.9% (Figure 1).

Figure 1: Dose rate profile at 2 mm in water and TG-149 consensus dose rate in water, normalized at z = 0 (left) and radial dose profile in water and TG-149 consensus dose rate in water, normalized at r = 2 mm (right).

The absorbed dose to the test patient was up to 77.9% lower in the entire artery segment and up to 50.5% lower in the target volume compared to the dose calculated in water (Figure 2). The dose distributions around the source in the patient-specific simulations were non-homogenous due to the guidewire, the stents and calcification.