Experimental and Monte Carlo-based determination of the beam quality specifier for TomoTherapyHD treatment units

Abstract Reference dosimetry by means of clinical linear accelerators in high-energy photon fields requires the determination of the beam quality specifier TPR 20,10 , which characterizes the relative particle flux density of the photon beam. The measurement of TPR 20,10 has to be performed in homogenous photon beams of size 10 × 10 cm 2 with a focus-detector distance of 100 cm. These requirements cannot be fulfilled by TomoTherapy treatment units from Accuray. The TomoTherapy unit provides a flattening-filter-free photon fan beam with a maximum field width of 40 cm and field lengths of 1.0 cm, 2.5 cm and 5.0 cm at a focus-isocenter distance of 85 cm. For the determination of the beam quality specifier from measurements under nonstandard reference conditions Sauer and Palmans proposed experiment-based fit functions. Moreover, Sauer recommends considering the impact of the flattening-filter-free beam on the measured data. To verify these fit functions, in the present study a Monte Carlo based model of the treatment head of a TomoTherapyHD unit was designed and commissioned with existing beam data of our clinical TomoTherapy machine. Depth dose curves and dose profiles were in agreement within 1.5% between experimental and Monte Carlo-based data. Based on the fit functions from Sauer and Palmans the beam quality specifier TPR 20,10 was determined from field sizes 5 × 5 cm 2 , 10 × 5 cm 2 , 20 × 5 cm 2 and 40 × 5 cm 2 based on dosimetric measurements and Monte Carlo simulations. The mean value from all experimental values of TPR 20,10 resulted in TPR 20 , 10 ¯ = 0.635 ± 0.4 % . The impact of the non-homogenous field due to the flattening-filter-free beam was negligible for field sizes below 20 × 5 cm 2 . The beam quality specifier calculated by Monte Carlo simulations was TPR 20,10  = 0.628 and TPR 20,10  = 0.631 for two different calculation methods. The stopping power ratio water-to-air s w , a Δ directly depends on the beam quality specifier. The value determined from all experimental TPR 20,10 data was s w , a Δ = 1.126 ± 0.1 % , which is in excellent agreement with the value directly calculated by Monte Carlo simulations. The agreement is a good indication that the equations proposed by Sauer and Palmans are able to calculate the beam quality specifier under reference conditions from measurements in arbitrary photon field sizes with high accuracy and are applicable for the TomoTherapyHD treatment unit.