Thesis Defence: Investigation of a Monte Carlo Model Based Dose Calculation for PD-103 Low-Dose Rate Breast Brachytherapy and I-125 Low-Dose Rate Prostate Brachytherapy

Dakota McKeown, supervised by Dr. Yiwen Xu and Dr. Christina Haston, will defend their thesis titled “Investigation of a Monte Carlo Model Based Dose Calculation for PD-103 Low-Dose Rate Breast Brachytherapy and I-125 Low-Dose Rate Prostate Brachytherapy” in partial fulfillment of the requirements for the degree of Master of Science in Medical Physics.

An abstract for Dakota McKeown’s thesis is included below.

Defences are open to all members of the campus community as well as the general public.  Registration is not required for in person defences.

---

ABSTRACT

Purpose: Clinical brachytherapy is typically performed using TG43 formalism, which assumes a homogeneous water medium and does not account for factors such as interseed attenuation or tissue inhomogeneities. Monte Carlo (MC) model-based dose calculation algorithms (MBDCA) provide improved dosimetric accuracy. The purpose of this work is to quantify the impact of eb_gui, a Canadian developed, user-friendly MC MBDCA, on plan evaluation with 80 low dose rate (LDR) breast brachytherapy patients and 76 LDR prostate brachytherapy patients.

Methods: Dosimetry of 80 breast brachytherapy cases treated with Pd-103 seeds (TheraSeed200, 2.3-2.8 U, 90 Gy prescription dose) and 76 prostate cases treated with I-125 seeds (AgX100, 0.354- 0.493 U, 144 or 110 Gy prescription doses) were simulated using the MC program and doses were calculated using two methods: 1) water-equivalent MC and 2) material-assigned MC. The clinical dose-volume histogram metrics of the target and organs at risk (OAR) for both simulations are extracted and compared across the two patient cohorts.

Results: Breast target volumes show a slight overestimation of dose by water-equivalent MC of 1% for the CTV V100 and 6% for the ETV V100 compared to the median material-assigned MC results (p < 0.05). Significant underestimation of dose is seen by water-equivalent MC results for the skin D0.2cm3 of 20.3% compared to material-assigned MC (p < 0.05). Water-equivalent MC for prostate target volume, CTV, show an overestimation of dose of 4.4% for the V100 compared to material-assigned MC (p < 0.05). The only potentially clinical significant result for a prostate OAR is the urethra D1cm3 showing an overestimation of dose of 6.3% by water-equivalent MC (p < 0.05).

Conclusion: MC simulation of 80 breast and 76 prostate LDR patients were performed. Slight overestimation of dose in target volumes and prostate OARs and potentially clinically significant underestimation of dose in skin were seen using the standard TG43 calculations when compared to the more accurate MBDCA. These differences motivate the future adoption of eb_gui as a valuable tool for LDR breast and prostate treatments in clinical practices.