Eric Shinohara MD, MSCI
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: March 18, 2009
Radiation is generally measured in units of absorbed dose (gray or rads) which are defined as the amount of energy absorbed per unit mass of tissue. However, this value does not take into account the differences in biological effectiveness between different types of radiation. The relative biological effectiveness (RBE) is a way to account for differences in biological efficacy between different types of radiation, using X-rays as the reference point. Prior studies have found the relative biological effectiveness of protons to be between 1.08 and 1.15. The RBE can be used to convert photon to proton dose. Proton dose is usually prescribed in cobalt gray equivalents (CGE). However, it is important to note that near the end of the Bragg peak there can be an increase in the RBE. The increased RBE near the end of the Bragg peak has been estimated to be as high as 2.05. A factor which is closely related to the RBE is linear energy transfer (LET). It is defined as the energy lost per unit distance by a particle as it travels along its track. Proton beams tend to be sparsely ionizing until they are near the end of their range. For an approximately 200 MeV beam, this corresponds to over 4 times more dose being deposited at the Bragg peak compared with the dose deposited at the area where the beam enters. The Oxygen Enhancement Ratio (OER) compares the amount of radiation required to give an effect in anoxic conditions versus the amount of radiation required in optimal oxygen conditions. Studies have demonstrated that there is no difference in OER between protons (2.5-3.0) and standard X-rays. Hence, the advantages that protons provide are mostly due to their physical properties rather than their biological effects on tissues.
The use of passive scatter proton beams creates increased levels of secondary neutron dose due to the protons' interactions with the scatter foils and collimating components of the nozzle. Specifically, the aperture used to shape the proton beam contributes the most secondary neutron contamination, as it is closest to the patient. Since the aperture is shaping the beam to a smaller size, the aperture is necessarily bombarded with a large amount of protons. Hence, the neutron dose can be limited by making sure that the smallest nozzle to match the aperture should be used. In this way, only the minimal amount of blocking will be used, minimize the neutron dose from the aperture. Additional moving the scatter foils further from the patient can also reduce neutron dose and is used in newer systems.However it is unclear what the clinical relevance of this neutron dose is. Part of the difficulty in determining the effect of secondary neutron production is the difficulty in measuring the whole body dose related to neutrons. These difficulties include: differences in measurement techniques used at different facilities and technical difficulties associated with the measurement of neutron dose in a mixed radiation field as well as within a phantom. These difficulties have provided a wide range of estimates for the whole body dose due to secondary neutrons. Nonetheless, it is generally agreed upon that the whole body neutron doses are too small to cause early or late radiation effects, but of primary concern is the risk of secondary malignancy. The best estimate for the RBE of low energy neutron is approximately 30. However, the majority of secondary neutrons produced in proton therapy will be high energy, and estimates of their RBE are in the range of 25. Calculations suggest that with passive scatter and scanning beam, the doses of secondary radiation are on the order of 10-2 Gy and 10-3 Gy per prescribed Gy of protons, respectively. Studies by Hall et al. have suggested that the lifetime risk for developing a malignancy for a 15 year-old treated with proton therapy with a passive scatter beam is 4.7% and 11.1% for a boy and girl, respectively. The risk was found to decrease substantially with age. It is important to note that these calculations are based on estimated exposures and RBE values. Hence, while it is unclear what the exact exposure and risks are in passive scatter proton beams, ways to minimize secondary neutron production should be implemented.
In an attempt to better answer the question of neutron dose and secondary malignancy, a retrospective matched study compared patients treated at the Harvard cyclotron facility from 1974-2001 with patients treated with photon therapy extracted from the SEER database. At total of 503 patients treated with protons were matched to one to three patients treated with radiation from the SEER database. Patients treated with a small proton field were not included in this study. Patients from age 1-90 were included in this study with the median age being 62. Median follow up for proton patients was 6.8 years versus 5.2 years for patients treated with photons. All patients were matched for age, year of treatment, histology and site of treatment. The primary outcome of this study was secondary malignancy. It is important to note that the majority of patients treated with protons also received photons for part of their treatment course (typically 20% of the treatment). The results from the study demonstrated that 32 patients (6.4%) treated with protons developed secondary malignancies as compared to 66 patients (13.1%), who received photons and that this difference was statistically significant. Scanning beams do not require the numerous components in the beam line that passive scatter beams require, and hence secondary neutron production is much less of an issue.
In addition to scattering, protons can also undergo nuclear reactions with a head-on collision with an atomic nucleus. These collisions can be elastic, with scatter of the proton or inelastic, resulting in a nuclear reaction. Nuclear reactions only make up a minority of the interactions that protons undergo while traveling through matter (about 20%). Near the surface, where the protons have the highest energy, the most likely nuclear interaction is an elastic collision between a proton and a nucleus in which the proton is scattered. However, the predominant proton interaction is still with atomic electrons and this interaction causes the majority of the proton energy loss and represented the dominant mechanism of energy loss in the region of the Bragg peak. Non-elastic proton collisions with the nucleus result in a nuclear reaction, producing energetic isotopes along with secondary protons, neutrons and occasionally heavy charged fragments, such as alpha particles. However, the resulting fluence from these particles is quite low and the majority of the dose delivered is attributed to the primary proton beam. A small number of positron-emitting isotopes are also produced, such as oxygen 16, which can produce an annihilation reaction resulting in gamma rays which can be detected. A PET scanner can be used both during proton therapy as well as after proton therapy to detect annihilation reactions.
Designing appropriate shielding for a proton facility has several aspects which are similar to designing shielding for a photon facility. The beam current, the work load of the room, the operating schedule, the amount of leakage and scatter radiation, the beam orientation factor, and area occupancy factor must all be accounted for in calculating appropriate shielding for protons as well as photons. Additionally, like photons, proton facilities must be equipped with radiation warning lights, monitors for patient observation, audio set up for communication with the patient, emergency shut off switches, and calibrated dosimetric devices for the safety of there personnel.
Shielding design for a proton facility must also take into account the production of secondary radiation dose, which is mainly comprised of neutrons, in addition to the primary proton beam. However, the literature on production of secondary neutrons in proton beams is sparse. Part of the reason for this is that secondary neutron production is dependent on the positioning of and the materials from which the beam delivery system is comprised of. There is no standard setup for the delivery system and hence it is difficult to translate the amount of secondary neutrons produced from one facility to another. Additionally, the protons can be given using either a passive scatter or a scanning beam technique, which affects secondary neutron production. The way in which the energy of the proton beam is modulated also determines the amount of secondary neutron produced. Synchrotrons are able to produce protons of various energies, whereas, cyclotrons require a range modulator, which generates more neutrons as it requires placing additional attenuating material within the beam.
The secondary radiation dose can be determined either through experiments or by calculations based on a model. As discussed above, experimental measurements are often difficult and the results are specific for the particular proton beam delivery system. Hence, it is not easy to extrapolate the experimental data to different treatment apparatuses. Calculations are also often complex and difficult. Monte Carlo calculations which model the beam delivery system can be used to try to estimate the dose delivered by secondary radiation, dose to the intended target, and dose to adjacent normal tissues. However this requires complex computations. The neutrons generated consist of evaporated neutrons which are released isotropically and high energy neutrons which are produced during intranuclear reactions, which are predominantly released in the forward direction. The intranuclear neutrons are a smaller fraction of the neutrons produced, but are the most penetrating component of secondary neutrons and must be accounted for.
The most common shielding used to attenuate secondary neutron production are earth, concrete, and steel. Steel has the advantage of being high density, creating a thinner shield; however, it also creates low energy neutrons and hence requires secondary shielding with concrete. The density and water content of the concrete affects neutron absorption and must be accounted for. The water content of concrete is generally thought to be about 5% and any deviation from this must be accounted for. Determining the shield transmission is also difficult and may be done using Monte Carlo models (such as FLUKA, LCS, and MARS) or experimental methods.
Another potential hazard for personnel and patients is induced radioactivity which can occur in any material which is irradiated by the primary beam or secondary radiation. The amount of induced radioactivity is dependent on the energy of the beam and type of material irradiated. Personnel may be exposed to induced radioactivity during the maintenance of the proton unit as well as when handling patient specific components, such as apertures. The appropriate monitoring and disposal of activated materials is crucial for the safety of the staff and patients. The accelerator and beam line may also be activated by the primary proton beam. Generally, aside from the targets and collimators, the activity of the activated components is fairly low and the majority of isotopes created are short lived.