Radioactive Beams for Image-Guided Particle Therapy: The BARB Experiment at GSI

Abstract

Several techniques are under development for image-guidance in particle therapy. Positron (β⁺) emission tomography (PET) is in use since many years, because accelerated ions generate positron-emitting isotopes by nuclear fragmentation in the human body. In heavy ion therapy, a major part of the PET signals is produced by β⁺-emitters generated via projectile fragmentation. A much higher intensity for the PET signal can be obtained using β⁺-radioactive beams directly for treatment. This idea has always been hampered by the low intensity of the secondary beams, produced by fragmentation of the primary, stable beams. With the intensity upgrade of the SIS-18 synchrotron and the isotopic separation with the fragment separator FRS in the FAIR-phase-0 in Darmstadt, it is now possible to reach radioactive ion beams with sufficient intensity to treat a tumor in small animals. This was the motivation of the BARB (Biomedical Applications of Radioactive ion Beams) experiment that is ongoing at GSI in Darmstadt. This paper will present the plans and instruments developed by the BARB collaboration for testing the use of radioactive beams in cancer therapy.

Keywords: PET; carbon ions; oxygen ions; particle therapy; radioactive ion beams.

Figure 1

Monte Carlo simulation of ¹²C and ¹¹C beams stopping in a spherical water volume and visualized by PET in 20 min. The graphs show the dose (red curve) and the activity (blue curve) distribution along the beam direction (z-axis) showing the shift between dose and activity when stable ions are used. Simulation by Monte Carlo code FLUKA.

Figure 2

Overview of the existing GSI-FAIR accelerator facility UNILAC-SIS-ESR. The FRS is the radioactive beam facility of GSI and a high-resolution magnetic spectrometer that provides a large variety of secondary nuclear beams ranging from hydrogen up to uranium, among them ^10,11C or ^14,15O, which can be produced from intense primary beams (like ¹²C or ¹⁶O) in the production target at its entrance. The ions of interest are identified in flight, spatially separated, energy bunched and used for experimental studies at the FRS itself (central and final focal plane, respectively, indicated by green dots) or they are transmitted via the FRS branch to the target hall for experiments in Cave M.

Figure 3

Schematic view of the symmetric FRS branch (projection in the horizontal dispersive plane) and the experimental setup to measure the interaction cross-section and nuclear charge composition of the beam fragmented in water. Red double-sided arrows indicate remotely removable detectors.

Figure 4

Schematic representation of the experimental setup and typical dose distributions measured with water column and the WERNER setup. (A) Technical drawing of the water column. (B) FLUKA simulation of laterally integrated 1D depth dose distribution for carbon ion isotopes of 250 MeV/u. (C) schematic representation of the WERNER setup. (D) 2D dose distribution map for a 4 cm- ¹²C SOBP.

Figure 5

The dual-panel UMCG positron imaging system installed at the Fragment Separator FRS at GSI. A beam of radioactive ions is coming from the right hand side and stopped in the PMMA phantom seen in the middle of the picture. The two detector panels are installed above and below the phantom each at a distance of 30 cm.

Figure 6

LMU hybrid γ-PET detector (A) A 3-layer PET detector developed at LMU Munich in collaboration with NIRS-QST. The PET detector consists of a 3-layer scintillator block, a light guide and an 8 × 8 SiPM array. (B) A flood map of the 3-layer PET detector exposed to a ²²Na radioactive point source.

Figure 7

The BARB4D setup: (1) SIRMIO cage with 32 PET detectors, (2) cylindrical phantom, (3) Compton camera, (4) Amplifier circuit boards for PET detectors, (5) translation stage, (6) rotating stage. The beam is entering the phantom from below. Right: Planned dose distribution for a static delivery with ellipsoidal target region.

Figure 8

Cellular phantom used for radiobiological measurements along a SOBP. The cells grow in monolayer on plastic plates that can be plunged at different position in the tank filled with water-equivalent growth medium. Plates are then removed after irradiation and the cell survival is measured in every position. In the BARB experiment, stable and radioactive carbon and oxygen isotopes will be used to irradiate CHO cells, under the null-hypothesis that no difference will be observed.

Figure 9

(A) Mouse CT image, coronal and sagittal planes, (B) example of an osteosarcoma in the C3H mouse hind limb; (C) the proposed hybrid Compton-PET scanner for the radioactive ion beam range verification, (D, E) the scanner configuration with the CT image of the mouse (sagittal and coronal planes are plotted, respectively) positioned along the scanner bore.

Figure 10

Impact of range margin reduction in robust optimization on the optic nerves sparing for a set of ACC patients. (A) Example of CT and two-field dose distribution for one of the analyzed patient plans. CTV, optic nerves, eyes and brainstem are contoured with black, white, purple and turquoise colors, respectively. Color bar represents the dose distribution (prescribed dose 3 Gy(RBE)). Primary goal of the robust optimization was achieving at least 95% of the prescribed dose in 95% of uncertainty scenarios (21 scenarios in total). Graphs represent sparing of left and right optic nerves in the analyzed plans with ¹²C beams (B) (3.5% range and 3 mm setup margin) and ¹¹C plans (C) (3 mm setup margin only), respectively. Each point corresponds to a single patient case (15 patients total).