Skip to ContentSkip to Navigation
KVI - Center for Advanced Radiation TechnologyOnderzoek en onderwijsMedical PhysicsResearch projects

In-vivo dose delivery verification

Full exploitation of the dosimetric benefits of protons requires in-vivo dose delivery verification to detect deviations from the treatment plan. Because in proton beam radiotherapy, the protons are stopped inside the patient body, in-vivo dose delivery verification requires imaging of secondary radiation created by the protons in the human body. High-energy photons are most suitable for this purpose: their interaction with the human tissue is rather weak and they are rather easily detected by suitable instruments. Two types of high-energy photons that follow from nuclear reactions induced by the protons are available:

  • positron annihilation photons (511 keV), following the decay of positron emitting isotopes
  • prompt gamma rays from the decay of excited nuclear energy levels

A good introduction to this subject is the review by K. Parodi (K. Parodi, “In vivo dose verification” in “Proton Therapy Physics”, Ed. H. Paganetti, CRC Press 2011, ISBN: 978-1-4398-3644-6).

The technology for imaging positron annihilation photons can be taken from the well-developed molecular imaging technique of Positron Emission Tomography (PET). Prompt-gamma-ray imaging is not unlike Single Photon Emission Computed Tomography (SPECT), another well-developed molecular imaging technique. However, standard equipment can not be used as the gamma-ray energies are much higher. SPECT uses photon energies of up to a few 100 keV whereas proton-induced prompt gamma rays have energies up to 7 MeV, with the major single line emissions at 6.13 MeV (from oxygen-16) and 4.44 MeV (from carbon-12).

Short-lived positron emitters: production and imaging

The major drawback of PET for dose delivery verification is that the information is delayed due to the radioactive decay. The most relevant positron emitter in a proton irradiation is oxygen-15 with a half-life of 2 minutes. So, to put it simply, a deviation from the treatment plan is detected 2 minutes after it occurs, a time comparable with the beam delivery. This delay also leaves room for biological washout, the phenomenon whereby radioactive isotopes travel through the body due to physiological processes such as blood flow.

Short-lived positron emitters obviously allow quicker feedback. With this in mind, we investigated the production of short-lived positron emitters in proton therapy. The results are available in: "Short-lived positron emitters in beam-on PET imaging during proton therapy", P. Dendooven et al., Phys. Med. Biol. 60(2015)8923, doi:10.1088/0031-9155/60/23/8923. The main conclusion is that nitrogen-12, with a half-life of just 11 ms, is a very promising positron emitter for near real time PET for dose delivery verification.

More recenlty, we have performed imaging of nitrogen-12 using a small PET system consiting of 2 modules each 65x65 mm2 with an efficiency of 0.27% in the center of the field-of-view. For 4000 detected nitrogen-12 decays, we show that a shift of 5 mm of the graphite target is detected as 6 ± 3 mm. Furthermore, a Monte Carlo simulation of a larger PET system shows that such a shift can be detected with millimeter accuracy: 5.5 ± 1.1 mm for 108 protons and 5.2 ± 0.5 mm for 5×108 protons. This makes fast and accurate feedback on the dose delivery during treatment possible. All details of this investigation are published in: "Beam-on imaging of short-lived positron emitters during proton therapy", H.J.T. Buitenhuis et al., Phys. Med. Biol. (2017) in press, doi:10.1088/1361-6560/aa6b8c.

Helium ion beam therapy

In recent years, virtually all particle beam treatments use either protons or carbon ions, with the ratio of number of proton to carbon treatments of about 6 (according to the PTCOG statistics for 2014, see www.ptcog.ch). The main rationale to use carbon ions instead of protons is the higher radiobiological effectiveness (RBE) in the Bragg peak region. However, the fragmentation of the carbon beam in the patient leads to a non-negligeable dose tail past the Bragg peak as well as a severe degradation of the Bragg peak for large ion ranges. The high LET of carbon ions in the plateau region before the Bragg peak and its effect on healthy tissue is also cause of concern, especially related to possible second cancers in pediatric patients. Also, carbon ion treatment per patient is more expensive due to the more expensive infrastructure needed. Proton beams do not create a dose tail and a proton therapy facility and thus the treatment per patient are less expensive. However, proton beams undergo a relatively large lateral spread; a clinically relevant issue for deeply penetrating proton beams.
These drawbacks of proton and carbon ion beams have recently generated a renewed interest in helium beam radiotherapy as it represents “the best of both worlds”: higher RBE and smaller lateral scattering with respect to protons and less fragmentation, a lower LET in healthy tissue and a less expensive treatment with respect to carbon ions. We are contributing to this renewed interest by investigating in-vivo dose delivery verification in helium ion therapy, looking both at positron emitters and prompt gamma rays.

Monte Carlo simulations

Our experimental work on in-vivo dose delivery verification is backed by Monte Carlo simulations. Next to simulating some of our experimental setups, we also simulate real patient treatments, calculating the production of secondary radiation, simulating the imaging and performing image reconstruction. This way the performance in the clinic of dose delivery verification techniques can be investigated.

Below some results from the simulations of a head-and-neck patient case. Shown are sagittal slices through a number of simulated distributions, overlayed on the treatment planning CT image.

Dose distribution from 1 field out of 3 in the treatment plan. The proton beam enters from the back of the head.
Dose distribution from 1 field out of 3 in the treatment plan. The proton beam enters from the back of the head.
Distribution of the production of the oxygen-15 positron emitter. Oxygen-15 is made in nuclear reactions on oxygen and is thus distributed rather smoothly throughout the irradiated region.
Distribution of the production of the oxygen-15 positron emitter. Oxygen-15 is made in nuclear reactions on oxygen and is thus distributed rather smoothly throughout the irradiated region.
Distribution of the production of the potassium-38 positron emitter. Potassium-38 is made in nuclear reactions on calcium and thus its distribution follows the skeleton.
Distribution of the production of the potassium-38 positron emitter. Potassium-38 is made in nuclear reactions on calcium and thus its distribution follows the skeleton.
Distribution of the production of 6.13 MeV prompt gamma rays produced on oxygen-16. As oxygen is distributed throughout the irradiated region, so is the production of this prompt gamma ray.
Distribution of the production of 6.13 MeV prompt gamma rays produced on oxygen-16. As oxygen is distributed throughout the irradiated region, so is the production of this prompt gamma ray.

Project members KVI-CART

Laatst gewijzigd:09 mei 2017 11:52