Influence of nozzle configurations on radiobiological effects in carbon ion radiation therapy estimated using GATE/Geant4.

Abstract

Background: Carbon Ion Radiation Therapy (CIRT) has been used in Europe and in Asia for several decades. The first treatment facility in America is now under construction at Mayo Clinic Florida (MCF) in Jacksonville. CIRT is known to have a variable relative biological effectiveness (RBE) which depends on the microdosimetric spectra and consequently the kinetic energy spectra.

Purpose: CIRT centers from different vendors exist around the world with different accelerators, delivery, and nozzle designs. Although nominally they provide comparable energies and beam qualities, this study investigates how the different nozzle designs might affect radiation quality and the consequent RBE.

Methods: The impact of three nozzle designs, the upcoming MCF (Jacksonville, USA), MedAustron Ion Therapy Center (Wiener Neustadt, Austria), and the Osaka HIMAK (Osaka, Japan), on the RBE was investigated using OpenGATE10. The microdosimetric spectra were determined using the abridged microdosimetric distribution methodology (AMDM) and input into the MCF microdosimetric kinetic model (MKM) to determine spectral and RBE dependence on nozzle design. Monoenergetic carbon ion beams having a range in water of 3 and 27 cm were simulated. For the 27 cm beams, a simulated water-based range shifter was inserted before the phantom, reducing the range to 3 cm. Furthermore, a shallow spread-out Bragg peak (SOBP) (5-10 cm) and a deep SOBP (15-21 cm), were simulated for all nozzles and the resulting integrated dose profiles compared.

Results: For all nozzle geometries, the range at 80% dose fall-off (R80) agreed within 0.1 mm. The lineal energy and the RBE agreed very well until the Bragg peak, after which some differences could be observed. For the SOBPs, the agreement was similar with an agreement in the biological dose before and at the SOBP within 0.7%. For the thick nozzle of Osaka HIMAK, small differences were observed, mostly in the fragmentation tail.

Conclusion: The AMDM was successfully integrated into OpenGATE10 and used to compute the RBE with the MCF MKM. It was shown that the nozzle design itself had only a minor effect on the radiation quality and consequently the RBE. A small difference in RBE is observed mostly after the Bragg peak and SOBP in the fragmentation tail and depends on the nozzle water equivalent tissue (WET), when it is a change of more than 24 mm.