Commissioning a spread-out Bragg peak (SOBP) delivery system is complicated, due to the variety in construction of such systems. Fully electromagnetic PBS systems based on unperturbed proton beams are, however, by definition all equal. Researchers at Massachusetts General Hospital (MGH) and Harvard Medical School (Boston MA) are exploiting this equivalence to simplify PBS commissioning.

The MGH team used Geant4 Monte Carlo calculations to generate a "Golden beam" data set of pristine Bragg peaks that can be applied to any PBS delivery system. Use of these data then reduces the commissioning task to the measurement of energy spread and initial phase space as a function of energy (Phys. Med. Biol. 57 1147).

"We are of the opinion that all fully electromagnetic PBS systems are equal. Unlike double scattering, a PBS treatment planning system (TPS) needs very few adjustable parameters to describe the dose deposited," explained lead author Benjamin Clasie. "Our goal is to reduce the number of adjustable parameters in the TPS to the smallest amount possible, make the TPS as independent of the machine as possible, and encapsulate the fundamental physics in the Golden beam data set."

Peak creation

Clasie and colleagues generated a set of pristine Bragg peaks in the clinical range of 20 to 350 mm (in water), with an energy spread of zero, and at peak-to-peak intervals of 5 mm. The peaks have units of Gy(RBE)mm2Gp–1, where Gp is 109 protons, thus providing a direct mapping from treatment planning parameters to integrated beam current.

The researchers used these peaks to calculate the expected absolute dose in the centre of a broad field, and then cross-calibrated Gy per Gp at three depths, by measuring dose with a Markus chamber and charge-per-spot with a Faraday cup. The measurements yielded a small correction factor that was applied to each generated pristine Bragg peak, yielding a set of experimentally verified Golden pristine Bragg peaks with a total error in the dose per Gp of about 4%.

The only facility-dependent calibration needed to describe depth-doses for a specific PBS delivery system is the energy spread of the mono-energetic proton beam, resulting from the accelerator and beam transport.

In this work, the researchers determine the energy spread by measuring proton pencil-beams with a PTW Bragg peak chamber. This device, with a diameter of 8.16 cm, requires approximately 5% correction for dose deposited outside its chamber, which the team determined using Monte Carlo simulations. The measured data are corrected and compared to the Golden pristine peak data to yield the energy spread.

"The energy spread can be determined through other means, eliminating the need for Bragg peak chamber measurements," Clasie noted. "It can be calculated from the configuration of the hardware, borrowed from facilities that share the same configuration, or determined through spectroscopic measurements."

The energy spread data are then convolved with Golden pristine peaks to yield a Bragg peak data set for a particular PBS delivery system.

Lateral dose distribution

The researchers also characterized the lateral spread of the primary beam as a function of energy of the pencil-beam emerging from the beam-line. This parameter is independent of the others and characteristic of the beam-line and accelerator.

Secondary contributions to the lateral dose distribution from scattered protons and secondary particles increase with depth and are universal. The main consequence is leakage of energy away from the primary beam. The researchers provide a parameterization of this contribution. The leaked energy is about 5% and must be included in dose calculations to allow sufficiently accurate absolute dose calculations.

Delivery system calibration

Separate to the treatment planning commissioning, the PBS delivery system must also be calibrated prior to treatment, in terms of the required range in the patient (controlled by the energy selection system) and the absolute number of protons (controlled by a reference ionization chamber). The reference ionization chamber is also calibrated in terms of MU (reference counts) per Gp as a function of proton energy, using a Faraday Cup. These calibrations are specific to each vendor, if not to each individual installation.

This stage completes the calibration of all the elements needed to deliver broad-field depth-dose distributions. The researchers used the data to construct SOBPs in the MGH treatment planning system, Astroid, and achieved a delivered dose agreeing to within ±2% with the planned dose.

The MGH team has incorporated the Golden beam data set into the Astroid planning system to automate the various options described in this paper. "The procedures in the paper will be used to simplify the TPS commissioning in our other treatment rooms," Clasie told medicalphysicsweb. "We also will ask other centres for their data and validate their dosimetry against this set as proof of its general applicability."

Related articles in PMB
Golden beam data for proton pencil-beam scanning
Benjamin Clasie et al Phys. Med. Biol. 57 1147
Range uncertainty in proton therapy due to variable biological effectiveness
Alejandro Carabe et al Phys. Med. Biol. 57 1159
Experimental characterization of lateral profiles of scanned proton and carbon ion pencil beams for improved beam models in ion therapy treatment planning
J Schwaab et al Phys. Med. Biol. 56 7813
A beam source model for scanned proton beams
Peter Kimstrand et al Phys. Med. Biol. 52 3151