KEY POINTS
- This Monte Carlo study used FLUKA to simulate monoenergetic proton beams at 60, 150, and 210 MeV and carbon-ion beams at 120, 280, and 400 MeV/u, together with a knife-edge collimator and pixelated lutetium-yttrium oxyorthosilicate detector.
- Prompt-gamma distal fall-off correlated with the Bragg peak for both particle types. Applying a 3-7 MeV energy window preserved fall-off position within 0.4 mm while reducing secondary-gamma contamination to approximately 1%-3% of the detected signal.
- Carbon ions produced higher prompt-gamma yields but substantially more neutrons. At 400 MeV/u, neutron yield reached 3.78692 per primary ion, approximately seven times the unfiltered prompt-gamma yield of 0.5431.
- Neutrons became the dominant detector background in carbon-ion simulations and generated lutetium-capture peaks at 5.8, 6.3, 6.5, and 6.6 MeV, flattening range-sensitive profiles and increasing statistical fluctuations.
- Under clinically relevant particle counts, estimated fall-off retrieval precision was approximately 0.2-0.7 mm for proton beams versus 2-6 mm for carbon-ion beams. The study used a homogeneous phantom, monoenergetic pencil beams, and idealized detector energy resolution.
CLINICAL TAKEAWAY
Prompt-gamma imaging remains a credible real-time range-verification approach for proton therapy, with the 3-7 MeV window offering a favorable balance between signal preservation and background suppression. Carbon-ion implementation is more difficult because clinically lower particle counts and substantially stronger neutron backgrounds reduce precision; neutron mitigation, detector optimization, and experimental validation remain necessary before clinical deployment.
SOURCE