Collimator design for spatially-fractionated proton beams for radiobiology research.

Preclinical and translational research is an imperative to improve the efficacy of proton radiotherapy. We present a feasible and practical method to produce spatially-modulated proton beams for cellular and small animal research for clinical and research facilities. The University of Washington (UW) 50.5 MeV proton research beamline hosting a brass collimation system was modeled using Monte Carlo simulations. This collimator consisted of an array of 2 cm long slits to cover an area of 2  ×  2 cm(2). To evaluate the collimator design effects on dose rate, valley dose and the peak-to-valley dose ratios (PVDR) the following parameters were varied; slit width (0.1-1.0 mm), peak center-to-center distance (1-3 mm), collimator thickness (1-7 cm) and collimator location along the beam axis. Several combinations of slit widths and 1 mm spacing achieved uniform dose at the Bragg peak while maintaining spatial modulation on the beam entrance. A more detailed analysis was carried out for the case of a slit width of 0.3 mm, peak center-to-center distance of 1 mm, a collimator thickness of 5 cm and with the collimator flush against the water phantom. The dose rate at 5 mm depth dropped relative to an open field by a factor of 12 and produced a PVDR of 10.1. Technical realization of proton mini-beams for radiobiology small animal research is demonstrated to be feasible. It is possible to obtain uniform dose at depth while maintaining reasonable modulation at shallower depths near the beam entrance. While collimator design is important the collimator location has a strong influence on the entrance region PVDRs and on dose rate. These findings are being used to manufacture a collimator for installation on the UW cyclotron proton beam nozzle. This collimator will enable comparative studies on the radiobiological efficacy of x-rays and proton beams.