Purpose: To analyze the impact of heterogeneity-corrected dose calculation on dosimetric quality parameters in gyne¬cological and breast brachytherapy using Acuros, a grid-based Boltzmann equation solver (GBBS), and to evaluate the shielding effects of different cervix brachytherapy applicators.

Material and methods: Calculations with TG-43 and Acuros were based on computed tomography (CT) retrospectively, for 10 cases of accelerated partial breast irradiation and 9 cervix cancer cases treated with tandem-ring applicators. Phantom CT-scans of different applicators (plastic and titanium) were acquired. For breast cases the V to lung, the D_0.1cm³, D_1cm³, D_2cm³ to rib, the D_0.1cm³, D_1cm³, D_10cm³ to skin, and D_max for all structures were reported. For cervix cases, the D_0.1cm³, D_2cm³ to bladder, rectum and sigmoid, and the D₅₀, D₉₀, D₉₈, V₁₀₀ for the CTV_HR were reported. For the phantom study, surrogates for target and organ at risk were created for a similar dose volume histogram (DVH) analysis. Absorbed dose and equivalent dose to 2 Gy fractionation (EQD2) were used for comparison.

Results: Calculations with TG-43 overestimated the dose for all dosimetric indices investigated. For breast, a decrease of ~8% was found for D_10cm³ to the skin and 5% for D_2cm³ to rib, resulting in a difference ~ –1.5 Gy EQD2 for overall treatment. Smaller effects were found for cervix cases with the plastic applicator, with up to –2% (–0.2 Gy EQD2) per fraction for organs at risk and –0.5% (–0.3 Gy EQD2) per fraction for CTV_HR. The shielding effect of the titanium applicator resulted in a decrease of 2% for D_2cm³ to the organ at risk versus 0.7% for plastic.

Conclusions: Lower doses were reported when calculating with Acuros compared to TG-43. Differences in dose parameters were larger in breast cases. A lower impact on clinical dose parameters was found for the cervix cases. Applicator material causes systematic shielding effects that can be taken into account.