Sairos Safai

Special report: Workshop on 4D-treatment planning in actively scanned particle therapy—Recommendations, technical challenges, and future research directions

This article reports on a 4D-treatment planning workshop (4DTPW), held on 7-8 December 2009 at the Paul Scherrer Institut (PSI) in Villigen, Switzerland. The participants were all members of institutions actively involved in particle therapy delivery and research. The purpose of the 4DTPW was to discuss current approaches, challenges, and future research directions in 4D-treatment planning in the context of actively scanned particle radiotherapy. Key aspects were addressed in plenary sessions, in which leaders of the field summarized the state-of-the-art. Each plenary session was followed by an extensive discussion. As a result, this article presents a summary of recommendations for the treatment of mobile targets (intrafractional changes) with actively scanned particles and a list of requirements to elaborate and apply these guidelines clinically.

Development of an inorganic scintillating mixture for proton beam verification dosimetry.

The availability at the Paul Scherrer Institute (PSI) of a spot-scanning technique with an isocentric beam delivery system (gantry) allows the realization of intensity-modulated proton therapy (IMPT). The development of 3D dosimetry is an important tool for the verification of IMPT therapy plans based on inhomogeneous 3D conformal dose distributions. For that purpose new dosimeters are being developed. The concept is to use a system of many millimetre sized scintillating volumes distributed in a polyethylene block, which are read on a CCD camera over a bundle of optical fibres and which can be irradiated from any direction orthogonal to the fibre axis. The purpose of this work is to investigate the composition of such small sensitive volumes. A mixture of inorganic phosphors and optical cement allows an optimal coupling between the scintillating volume and the optical fibre. Five different inorganic phosphors, available as powder, have been examined by considering their response along the Bragg curve. In particular, two phosphors have shown interesting behaviours: Gd2O2S:Tb and (Zn, Cd)S:Ag. Both phosphors have a high emission efficiency but contrasting behaviour in the Bragg peak region. The efficiency of Gd2O2S:Tb decreases with increasing stopping power (quenching of luminescence) while that of (Zn, Cd)S:Ag increases. Because of these contrasting behaviours it is possible to prepare a mixture of the two scintillating powders in a certain ratio in order to modulate the height of the measured Bragg peak relative to the entrance value so that it is in agreement with the ionization chamber measurements. We propose to use a mixture for the sensitive volume consisting of the following weight fractions: 48% Gd2O2S:Tb, 12% (Zn, Cd)S:Ag and 40% optical cement.

Improving the precision and performance of proton pencil beam scanning

In this report we present the technical features of Gantry 2, the new second generation scanning system of PSI. On the basis of the experience and success with the first prototype, Gantry 1, built in the 90s for introducing pencil beam scanning and IMPT into the field of proton therapy, we have recently implemented a new system capable of offering much faster repainted conformal scanning for being able to treat moving targets with scanning under image guidance, the next challenge in the field of proton therapy. The new technical developments are conducted in parallel to the ongoing basic commissioning of Gantry 2, which should go into operation with usual discrete spot scanning for treating static targets in 2013. The innovative layout of Gantry 2 and the integration in the treatment area of the basic equipment for image guidance are presented. Noteworthy are the sliding CT within reach of the patient table and the unique new Beam’s-Eye-View X-ray fluoroscopy system for taking images in the beam direction synchronized with the proton beam delivery. The first preliminary results with the development of much faster scanning modes look very encouraging. We can change the beam energy with the beam line within 80 ms for typical 0.5 cm range steps. We can deliver whole fluence-shaped energy layers within a time of the order of 100 ms. Dose lines are painted by changing the velocity of the scan magnets. The instantaneous dose rate of the pencil beam can be varied dynamically as well. The dose is precisely controlled with a feedback loop connecting the main gantry beam monitor with a vertical deflector plate at the ion source. These new fast scanning modes should be used for providing scanning with repainting, gating and tracking for treating moving targets. The goal is to develop pencil beam scanning as a universal beam delivery solution capable of treating optimally all possible clinical indications for proton therapy. Scanning could then completely replace the old beam delivery methods based on passive scattering from the market. The long term projects of Gantry 2 should represent the new contributions of PSI to the proton therapy field in the next 5-10 years, by providing direct translational cancer research from the physics laboratory into industry and clinics.

Independent dose calculations for commissioning, quality assurance and dose reconstruction of PBS proton therapy

Pencil beam scanning proton therapy allows the delivery of highly conformal dose distributions by delivering several thousand pencil beams. These beams have to be individually optimised and accurately delivered requiring a significant quality assurance workload. In this work we describe a toolkit for independent dose calculations developed at Paul Scherrer Institut which allows for dose reconstructions at several points in the treatment workflow. Quality assurance based on reconstructed dose distributions was shown to be favourable to pencil beam by pencil beam comparisons for the detection of delivery uncertainties and estimation of their effects. Furthermore the dose reconstructions were shown to have a sensitivity of the order of or higher than the measurements currently employed in the clinical verification procedures. The design of the independent dose calculation tool allows for a high modifiability of the dose calculation parameters (e.g. depth dose profiles, angular spatial distributions) allowing for a safe environment outside of the clinical treatment planning system for investigating the effect of such parameters on the resulting dose distributions and thus distinguishing between different contributions to measured dose deviations. The presented system could potentially reduce the amount of patient-specific quality assurance measurements which currently constitute a bottleneck in the clinical workflow.

Proton beam monitor chamber calibration.

The first goal of this paper is to clarify the reference conditions for the reference dosimetry of clinical proton beams. A clear distinction is made between proton beam delivery systems which should be calibrated with a spread-out Bragg peak field and those that should be calibrated with a (pseudo-)monoenergetic proton beam. For the latter, this paper also compares two independent dosimetry techniques to calibrate the beam monitor chambers: absolute dosimetry (of the number of protons exiting the nozzle) with a Faraday cup and reference dosimetry (i.e. determination of the absorbed dose to water under IAEA TRS-398 reference conditions) with an ionization chamber. To compare the two techniques, Monte Carlo simulations were performed to convert dose-to-water to proton fluence. A good agreement was found between the Faraday cup technique and the reference dosimetry with a plane-parallel ionization chamber. The differences-of the order of 3%-were found to be within the uncertainty of the comparison. For cylindrical ionization chambers, however, the agreement was only possible when positioning the effective point of measurement of the chamber at the reference measurement depth-i.e. not complying with IAEA TRS-398 recommendations. In conclusion, for cylindrical ionization chambers, IAEA TRS-398 reference conditions for monoenergetic proton beams led to a systematic error in the determination of the absorbed dose to water, especially relevant for low-energy proton beams. To overcome this problem, the effective point of measurement of cylindrical ionization chambers should be taken into account when positioning the reference point of the chamber. Within the current IAEA TRS-398 recommendations, it seems advisable to use plane-parallel ionization chambers-rather than cylindrical chambers-for the reference dosimetry of pseudo-monoenergetic proton beams.

Preliminary investigations for the option to use fast uniform scanning with compensators on a gantry designed for IMPT

PURPOSE In this experimental study, the authors explored the possibility to deliver the dose for proton therapy with fast uniform scanning on a gantry primarily designed for the delivery of conformal beam scanning and IMPT. The uniform scanning submode has been realized without equipment modifications by using the same small pencil beam used for conformal scanning, resulting in reduced realization costs. Uniform scanning has recently been adopted in a few proton therapy centers, as a basic beam delivery solution, and as an alternative to the use of scattering foils. The option to use such a mode to mimic scattering on a full-fledged scanning gantry could be of interest for treating some specific indications and as a possible solution for treating moving targets. METHODS Uniform iso-energy dose layers were painted by fast magnetic scanning alternated with fast energy changes with the gantry beam line. The layers were stacked and repainted appropriately to produce homogeneous three-dimensional dose distributions. A collimator∕compensator was used to adjust the dose to coincide laterally∕distally with the target volume. In addition, they applied volumetric repainting, since they are confident that this will further mitigate the effects of organ motion as compared with the presently used clinical scanning solutions. With the approach presented in this paper, they can profit from the higher flexibility of the scanning system to obtain additional advantages. For instance the shape of the energy layers can be adjusted to the projected target shape in order to reduce treatment time and neutrons produced in the collimator. The shape of the proximal layers can be shrunk, according to the cross section of the target at the corresponding range. This provides variable range modulation (proximal conformity) while standard scattering only provides fixed range modulation with unnecessary 100% dose proximal to the target. The field-specific hardware for a spherical target volume was mounted on the Gantry 2 nozzle. One field with proximal field size shrinking and one without, each of 1 Gy, were delivered. The dose distributions at different depths were recorded as CCD images of a scintillating screen. RESULTS The time to scan the volume once was about 4 s and the total delivery time was approximately 30 s. For the field with proximal conformity, dose sparing of up to 25% was measured in the region proximal to the target. A repainting capability of 48 times was achieved on the most distal layer. The proximal layers were repainted more due to the contribution of the plateau dose from the deeper layers. CONCLUSIONS The flexibility of a fast scanning gantry with very fast energy changes can easily provide beam delivery by uniform layer stacking with a significant degree of volumetric repainting and with the benefit of a dose reduction proximal to the target volume.

SU-D-BRE-01: A Realistic Breathing Phantom of the Thorax for Testing New Motion Mitigation Techniques with Scanning Proton Therapy

PURPOSE A prototype breathing phantom (named LuCa) has been developed which simulates the anatomy and motion of a patient thorax.In this work, we describe the results of the first commissioning tests with LuCa. METHODS The phantom provides a close representation of the human thorax. The lungs,contained within a tissue-equivalent ribcage and skin,are made from a polymer foam,which is inflated and deflated using a custommade ventilator. A tumor is simulated using a wooden ball with cutplanes for placing GafChromic films. The ventilator,controlled with Labview software,simulates a full range of breathing motion types.Commissioning tests were performed to assess its performance using imaging (CT and radiographic) and film dosimetry as follows:i)maximum Tumor excursion at acceptable pressure ranges, ii)tumor Motion repeatability between breathing periods,iii)reproducibility between measurement days,iv)tumor-to-surface motion correlation and v)reproducibility of film positioning in phantom. RESULTS The phantom can generate repeatable motion patterns with sin4 ,sin,breath-hold (tumor amplitude repeatability <0.5mm over 10min),aswell as patient-specific motion types. Maximum excursions of the tumor are 20mm and 14mm for the large and small tumor inserts respectively. Amplitude reproducibility (Coefficient of Variation) averaged at 16% for the workable pressure range over 2 months. Good correlation between tumor and surface motion was found with R2 =0.92. Reproducibility of film positioning within the thorax was within 0.9mm, and maximum 3° error from the coronal plane. Film measurements revealed that the film repositioning error yields relative errors in the mean dose over the planned target volume (PTV) of up to 2.5% and 4.5% for films at the center and on the edge of the PTV respectively. CONCLUSION Commissioning tests have shown that the LuCa phantom can produce tumor motion with excellent repeatability. However,a poorer performance in reproducibility of tumor amplitude for a given peak pressure week-to-week. Film set-up reproducibility is adequate for detection of dosimetric errors resulting from motion of >3%. This work is funded by Swiss National Fund Grants 320030_127569 and 320030_1493942-1.

The role of a microDiamond detector in the dosimetry of proton pencil beams.

In this work, the performance of a microDiamond detector in a scanned proton beam is studied and its potential role in the dosimetric characterization of proton pencil beams is assessed. The linearity of the detector response with the absorbed dose and the dependence on the dose-rate were tested. The depth-dose curve and the lateral dose profiles of a proton pencil beam were measured and compared to reference data. The feasibility of calibrating the beam monitor chamber with a microDiamond detector was also studied. It was found the detector reading is linear with the absorbed dose to water (down to few cGy) and the detector response is independent of both the dose-rate (up to few Gy/s) and the proton beam energy (within the whole clinically-relevant energy range). The detector showed a good performance in depth-dose curve and lateral dose profile measurements; and it might even be used to calibrate the beam monitor chambers-provided it is cross-calibrated against a reference ionization chamber. In conclusion, the microDiamond detector was proved capable of performing an accurate dosimetric characterization of proton pencil beams.

A study of lateral fall-off (penumbra) optimisation for pencil beam scanning (PBS) proton therapy

The lateral fall-off is crucial for sparing organs at risk in proton therapy. It is therefore of high importance to minimize the penumbra for PBS. Three optimisation approaches are investigated: Edge-collimated uniformly weighted spots (collimation), pencil beam optimisation of uncollimated pencil beams (edge-enhancement) and the optimisation of edge collimated pencil beams (collimated edge-enhancement). To deliver energies below 70MeV, these strategies are evaluated in combination with the following pre-absorber methods: field specific fixed thickness pre-absorption (fixed), range specific, fixed thickness pre-absorption (automatic) and range specific, variable thickness pre-absorption (variable). All techniques are evaluated by Monte Carlo simulating square fields in a water tank. For a typical air gap of 10cm, without pre-absorber collimation reduces the penumbra only for water equivalent ranges (WER) between 4cm to 11cm by up to 2.2mm. The sharpest lateral fall-off is achieved through collimated edge-enhancement, which lowers the penumbra down to 2.8mm. When using a pre-absorber, the sharpest fall-offs are obtained when combining collimated edge-enhancement with a variable pre-absorber. For edge-enhancement and large air gaps, it is crucial to minimize the amount of material in the beam. For small air gaps however, the superior phase space of higher energetic beams can be employed when more material is used. In conclusion, collimated edge-enhancement combined with the variable pre-absorber is the recommended setting to minimize the lateral penumbra for PBS. Without collimator, it would be favourable to use a variable pre-absorber for large air gaps and an automatic pre-absorber for small air gaps.The lateral fall-off is crucial for sparing organs at risk in proton therapy. It is therefore of high importance to minimize the penumbra for pencil beam scanning (PBS). Three optimisation approaches are investigated: edge-collimated uniformly weighted spots (collimation), pencil beam optimisation of uncollimated pencil beams (edge-enhancement) and the optimisation of edge collimated pencil beams (collimated edge-enhancement). To deliver energies below 70 MeV, these strategies are evaluated in combination with the following pre-absorber methods: field specific fixed thickness pre-absorption (fixed), range specific, fixed thickness pre-absorption (automatic) and range specific, variable thickness pre-absorption (variable). All techniques are evaluated by Monte Carlo simulated square fields in a water tank. For a typical air gap of 10 cm, without pre-absorber collimation reduces the penumbra only for water equivalent ranges between 4-11 cm by up to 2.2 mm. The sharpest lateral fall-off is achieved through collimated edge-enhancement, which lowers the penumbra down to 2.8 mm. When using a pre-absorber, the sharpest fall-offs are obtained when combining collimated edge-enhancement with a variable pre-absorber. For edge-enhancement and large air gaps, it is crucial to minimize the amount of material in the beam. For small air gaps however, the superior phase space of higher energetic beams can be employed when more material is used. In conclusion, collimated edge-enhancement combined with the variable pre-absorber is the recommended setting to minimize the lateral penumbra for PBS. Without collimator, it would be favourable to use a variable pre-absorber for large air gaps and an automatic pre-absorber for small air gaps.

Experimental validation of beam quality correction factors for proton beams

This paper presents a method to experimentally validate the beam quality correction factors (kQ) tabulated in IAEA TRS-398 for proton beams and to determine the kQ of non-tabulated ionization chambers (based on the already tabulated values). The method is based exclusively on ionometry and it consists in comparing the reading of two ionization chambers under the same reference conditions in a proton beam quality Q and a reference beam quality (60)Co. This allows one to experimentally determine the ratio between the kQ of the two ionization chambers. In this work, 7 different ionization chamber models were irradiated under the IAEA TRS-398 reference conditions for (60)Co beams and proton beams. For the latter, the reference conditions for both modulated beams (spread-out Bragg peak field) and monoenergetic beams (pseudo-monoenergetic field) were studied. For monoenergetic beams, it was found that the experimental kQ values obtained for plane-parallel chambers are consistent with the values tabulated in IAEA TRS-398; whereas the kQ values obtained for cylindrical chambers are not consistent--being higher than the tabulated values. These results support the suggestion (of previous publications) that the IAEA TRS-398 reference conditions for monoenergetic proton beams should be revised so that the effective point of measurement of cylindrical ionization chambers is taken into account when positioning the reference point of the chamber at the reference depth. For modulated proton beams, the tabulated kQ values of all the ionization chambers studied in this work were found to be consistent with each other--except for the IBA FC65-G, whose experimental kQ value was found to be 0.6% lower than the tabulated one. The kQ of the PTW Advanced Markus chamber, which is not tabulated in IAEA TRS-398, was found to be 0.997 ± 0.042 (k = 2), based on the tabulated value of the PTW Markus chamber.