Search

Particle therapy (PT) used for cancer treatment can spare healthy tissue and reduce treatment toxicity. However, full exploitation of the dosimetric advantages of PT is not yet possible due to range uncertainties, warranting development of range-monitoring techniques. This study proposes a novel range-monitoring technique introducing the yet unexplored concept of simultaneous detection and imaging of fast neutrons and prompt-gamma rays produced in beam-tissue interactions. A quasi-monolithic organic detector array is proposed, and its feasibility for detecting range shifts in the context of proton therapy is explored through Monte Carlo simulations of realistic patient models and detector resolution effects. The results indicate that range shifts of 1mm can be detected at relatively low proton intensities (22.30 (13) × 10 7 protons/spot) when spatial information obtained through imaging of both particle species are used simultaneously. This study lays the foundation for multi-particle detection and imaging systems in the context of range verification in PT. © 2023, The Author(s).

Due to its high sensitivity to anatomical changes, particle therapy will only unfold its full potential together with a functioning online range verification. We present a detector concept making use of a large fraction of the secondary particles available by hybrid prompt gamma-ray and fast neutron imaging. The system is expected to exhibit a high detection sensitivity to these particles, a high time, energy and position resolution, excellent pulse shape discrimination, and a small footprint. It comprises a quasi-monolithic organic detector array consisting of novel organic scintillators with dual-ended silicon photomultiplier light read-out and fast digitizers. The reconstruction of the proton range from the events registered by the detector is based on gamma/neutron scatter kinematics, cone back-projection and maximum likelihood expectation maximization. Multiple studies are currently ongoing investigating the feasibility of this concept on an experimental and simulation level. A first Monte Carlo simulation study involving realistic patient data and an idealized detector has revealed that a range-shift sensitivity of 1 mm per spot is attainable for clinical spot weights. These results demonstrate the potential of particle treatment verification by fast neutrons and prompt gamma-rays and strengthen the potential of this hybrid system for clinical application.

Introduction Particle therapy emerged as a principal innovative technology for tumor treatment. However, it requires verification to exploit its full potential. Various approaches based on prompt gamma radiation were developed for range verification. Prompt gamma-ray timing (PGT), which determines the range of the therapeutic particles from time distributions of produced secondary gamma-rays, is a promising candidate for this as it is collimator-free and can be easily integrated into existing clinical beam deliveries [1]. However, phase instabilities between the accelerating radio frequency (RF) and the actual proton arrival time make phase monitoring indispensable [2]. In recent years, different concepts for proton bunch monitors (PBM) were developed [3,4]. The two most promising candidates are presented here. Materials and Methods A diamond detector with high radiation hardness and excellent time resolution was positioned close to the beam degrader. Protons scattered there were used to determine the phase correlation between RF and proton arrival time under realistic clinical conditions. The second monitor is a direct phasing tap on the low-level RF module of the IBA Proteus®C230 isochronous cyclotron. The correlation of both PBMs to the actual phase was checked in a scattering setup where protons from a pencil beam scattered in a thin polythene foil and then were coincidentally detected by CeBr₃-scintillators. The kinematics of this reaction makes the determination of the proton arrival time independent of the beam's bunch time spread. Results Both PBMs strongly correlate with the phase shift between the arrival time of the proton bunch and the RF. A model based on an over-damped, harmonic oscillator was able to describe the time-dependent change in the phase position and its correction by a control circuit with sufficient accuracy. Thus, the phase instabilities in the PGT data could be successfully corrected, enabling the improved measurement accuracy. Summary The phase instability of the proton bunch is the source of the greatest uncertainty of the PGT method [2]. Two PBMs were studied to correct these instabilities. PBM(s) complement the measurement setup, which should increase the sensitivity of the method and facilitate the translation of the PGT approach into clinical application. Literature [1] C. Golnik et al.: Range assessment in particle therapy based on prompt γ-ray timing measurements, Phys. Med. Biol. 59 (2014) 5399. [2] T. Werner et al.: Processing of prompt gamma-ray timing data for proton range measurements at a clinical beam delivery, Phys. Med. Biol. 64 (2019) 105023. [3] F. Permatasari: Development of a Clinically Applicable Technique for Range Verification in Proton Therapy Based on the PGT Method, PhD thesis, TU Dresden, in preparation. [4] R.-D. Werner: Charakterisierung eines schnellen Diamantdetektors als PBM für die Reichweiteverifikation in der Protonentherapie, Masterarbeit, Universität Halle-Wittenberg, 2021.

Range verification is an important prerequisite to unfold the full potential of the finite range of proton beams and to improve treatment precision. The prompt gamma-ray timing (PGT) technique offers a non-invasive approach for range verification using the measured time distribution of the prompt gamma rays produced in the patient. PGT dispenses with a heavy collimator and can be integrated into existing treatment gantries. However, the high sensitivity of this technique to any instabilities in the proton bunch periodicity is a major challenge and demands online monitoring of the proton bunch arrival time. Therefore, we have developed a proton bunch monitor (PBM) comprising fast-scintillating fibers with a double-sided silicon photomultiplier readout. Placing the PBM in the beam halo allows the direct measurement of the proton arrival time at clinical beam intensities while maintaining a processable trigger rate. In a proof-of-principle experiment with a thick acrylic glass target and defined cylindrical air cavities as well as tissue equivalent inserts, a direct monitoring of proton bunches was carried out together with a PGT measurement. With the use of the PBM, another important step towards the clinical translation of the PGT method was taken.

Treatment verification is expected to improve targeting precision in particle therapy. A promising technique to achieve this goal is the detection of prompt gamma rays emitted along the particle tracks inside the patient. The range of the particle beam can be inferred by determining the time distribution of these gamma-rays relative to the radio frequency of the accelerator, a method commonly referred to as Prompt Gamma-Ray Timing. However, the translation of this method into a clinical setting is currently hindered by instabilities of the phase relation between the arrival of the proton bunches and the radio frequency of the accelerator. These instabilities include two effects, which have been studied at the clinical treatment facility of the University Proton Therapy Dresden. Firstly, a long-term drift of the proton bunch phase relative to the radio frequency in the order of several hundred picoseconds per hour was observed, which may be caused by small temperature changes in the cyclotron’s magnet resulting in magnetization variations in its iron parts. Secondly, strongly damped oscillations in the mean of measured prompt gamma-ray timing spectra with an amplitude in the order of few hundred picoseconds occur for about two seconds after each change of the particle energy during pencil beam scanning. This oscillation is caused by ramping the acceleration voltage back to its nominal value, which is reduced between energy layers to minimize the dark current of the accelerator and the resulting excess dose to the patient. While the former effect is only of secondary importance for the treatment due to its comparably long time scale, the phase oscillation has a considerable negative impact on the accuracy of the Prompt Gamma-Ray Timing method, which has to detect time shifts in the order of a few picoseconds for the detection of millimeter range changes. Therefore, the development of a method to monitor the arrival time of the proton bunches independently from the accelerator radio frequency, a so-called proton bunch monitor, is crucial. To this end, a bunch monitor prototype was developed consisting of scintillating fibers placed in the halo of the proton beam. The fibers were read out on both ends by silicon photomultipliers. A thick acrylic glass target with cylindrical air cavities of varying thickness and different tissue-equivalent inserts was irradiated with protons of clinically relevant energies and typical beam currents. The mean proton arrival time, determined from the time spectra of the proton bunch monitor, was used to correct the prompt gamma-ray timing spectra acquired by Ø2”x2”CeBr3 high-resolution scintillation detectors. This correction allowed to resolve differences in the prompt-gamma ray timing spectra acquired with the different cavities and inserts. In conclusion, the developed proton bunch monitor was successfully integrated to the Prompt Gamma-Ray Timing method and is expected to enable the clinical application of this method for clinical treatment verification in particle therapy.

Particle therapy is a modality for treating cancer using ionizing radiation from, e.g., protons or carbon ions. A growing number of patients worldwide receive particle therapy as a part of their cancer treatment due to its dosimetric advantages over the more conventional external beam radiotherapy using photons. The finite range of particles in tissue results in sparing of healthy tissue surrounding the tumour and thereby a reduced risk of adverse effects compared to conventional treatment. This opens possibilities for a more intensified treatment by dose escalation to the tumour and thereby an increased probability for controlling the disease. However, charged particles, when used for external beam radiotherapy, are much less forgiving than photons in case of treatment deviations. Treatment deviations may be caused by (1) sensitivity of charged particles to anatomical or density changes along their radiological path in the patient, e.g., due to organ motion or tumour shrinkage and (2) uncertainties associated with the estimation of the exact position of the Bragg-peak (maximum dose deposition) in tissue, e.g., due to uncertainties in the estimation of stopping power ratios. Collectively, these are referred to as “range uncertainties”. The most important consequence of range uncertainties is that the tissue sparing potential of charged particles cannot be fully exploited. Therefore, there is an urgent need for reliable and robust treatment verification systems that can identify potential treatment deviations in real-time. Despite recent advances in state-of-the-art prompt gamma-ray imaging, spectroscopy and timing systems as well as in-beam and offline PET imaging systems, the development of treatment verification systems still lags and as such, there is no system yet in wide routine clinical use. Common to all existing solutions is the fact that they rely on a single feature of a single particle species and will therefore suffer from limited counting statistics. This is an important limiting factor in applying state-of-the-art systems in a treatment verification system. Recently, a collaboration of detector, nuclear and medical physicists, nuclear engineers and mathematicians initiated the NOVO (Neutron and gamma-ray imaging with quasi-monolithic organic detector arrays – a novel, holistic approach to real-time range assessment-based treatment verification in particle therapy) project. The aim of the NOVO project is to develop a sophisticated quasi-monolithic organic detector array (QuDA) that will combine detection and imaging of secondary fast neutrons and prompt gamma-rays produced in nuclear interactions of incident particles with tissue, the profiles of which show a strong correlation with the incident particle beam range. In addition to imaging spatial profiles of secondary fast neutrons and prompt gammarays, the QuDA will be used to acquire information on additional features of each particle species, such as timing, energy and intensity, that can provide supplementary information on the range of particle beams in tissue, thus representing a major shift from single-feature, single-particle systems to a multi-feature, multiparticle system. In this contribution we will report on first estimates of the imaging properties of a potential QuDA design, based on Monte Carlo radiation transport models with MCNP6.2, Geant4 and GATE, considering fast neutrons and prompt gamma-rays for various, realistic combinations of timing, energy and position resolution of the individual sensing elements. Furthermore, we will report on the expected detection efficiencies and the resulting range monitoring precision for proton beams with clinically relevant energies and intensities incident on homogeneous polymethyl methacrylate phantoms. The preliminary results demonstrate the potential to obtain a range monitoring precision of approximately 1 mm down to proton beam intensities of about 1 – 2 x 10⁷ [protons/pencil beam] when fast neutron and prompt gamma-ray data are combined.

Particle therapy constitutes a promising and rapidly developing method in modern cancer treatment. In order to exploit its full potential, however, it requires detailed dose verification. Although the applicability of in-beam positron emission tomography and prompt gamma rays has already been demonstrated in patients, range verification is not yet part of the clinical routine in particle therapy. This is due to not only the methodological limitations of previous systems, but also to commercial, clinical and physical boundary conditions. In pencil beam scanning, the state-of-the-art treatment method in particle therapy, the number of secondary particles (essentially positrons, prompt gamma rays and fast neutrons) available per spot (Δt = 10 ms to 100 ms) is limited. This leads to statistical accuracy limits for verification systems exploiting these secondary particles as range probes. The development of a clinically useable treatment verification system requires gathering as much information about the local dose, as possible. The instantaneous fluence rate of prompt gamma rays reaching 5×10⁶ cm‾²s‾¹ to 10⁸ cm‾²s‾¹ challenges modern data acquisitions connected to monolithic inorganic scintillators with typical sizes used in present verification systems. In order to reduce the detector load, and also with regard to the ever higher intensities of next generation medical accelerators, future systems have to be more granular. Multi-Feature Treatment Verification combines and extends established methods (prompt gamma-ray imaging, spectroscopy, timing, etc.) in order to achieve higher reliability and performance. This idea was taken up by the NOVO project and expanded by a multi-particle approach. The NOVO (i.e. N eutron and gamma ray imaging with quasi-monolithic organic detector arrays – a novel, holistic approach to real-time range assessment-based treatment verification in particle therapy) consortium is a large collaboration of medical, nuclear and detector physicists, nuclear engineers, and mathematicians, which aim to develop a holistic realtime treatment verification system in particle therapy. Elements of a potential multi-feature/multi-particle treatment verification multi-channel system were characterized in a double time-of-flight experiment at the pulsed photo-neutron source nELBE (neutrons @ Electron Linac for beams with high Brilliance and low Emittance). The essential properties (time resolution, light yield, detection efficiency and pulse shape discrimination) of an EJ-276 plastic scintillator were determined. The very first experimental results show that the time resolution (ΔT < 400 ns) of a 20 × 20 × 200 mm³ EJ-276 plastic scintillator with double-sided readout will reach the high demands of such a proposed range verification system. However, the determined quality of the pulse-shape discrimination, the energy resolution and the quite high neutron detection threshold of above 200 keV show that the light yield of this type of scintillator is not high enough to be used in a multi-feature-based treatment verification system. Particle transport calculations with MCNP6 and GEANT4 were performed to confirm the experimental results of a single detector element. Furthermore, they also show a promising measurement accuracy of a multi-channel overall system.

Objective: To improve the accuracy of range verification with prompt-gamma-ray imaging (PGI), enabling the validation of CT-based stopping-power prediction in patients. Material & Methods: A PGI-slit-camera system was modified to enhance its positioning accuracy, now using a floor-based docking station. The camera position is calibrated with orthogonal X-rays and its reproducibility was validated with X-ray measurements at two different days with ten repositioning iterations each. To determine the PGI simulation accuracy, the camera position derived with the X-Ray system and PGI-based range shift determination in a PMMA phantom (measured vs. simulated PGI profiles) was correlated. Subsequently, the PGI system was clinically applied to monitor absolute proton ranges for a 1.5Gy field during eight fractions of a hypo-fractionated prostate-cancer treatment using pencil beam scanning (Fig.1). For all monitored fractions, in-room control CT scans were acquired in treatment position, enabling PGI-based range analysis for the actual patient anatomy. Results: The reproducibility of the camera position in beam direction was ±0.55mm (1σ) over different days. A 1.1mm offset in absolute range determination was found. It can be directly identified as simulation accuracy and is corrected in subsequent clinical application. The overall PGI range measurement uncertainty of about 2mm (averaging over multiple spots for global-shift determination) is well below the range prediction uncertainty (3.5%∙Range+2mm). Evaluation of the clinical slit-camera application and the verification of the applied stopping-power prediction using dual-energy CT is ongoing. Conclusion: The accuracy of PGI-based range verification was improved to enable the verification of CT-based stopping-power prediction in patients, potentially allowing for a future reduction of currently used range uncertainties.

Purpose/ Objective In-vivo prompt-gamma imaging (PGI) is a promising method for directly assessing deviations in the proton range during proton therapy. However, several effects that can cause range shifts in patients need to be distinguished, e.g. global errors in CT conversion to stopping power ratio (SPR), variations in patient setup, and changes in the patient anatomy. Here, we evaluate if the source of range deviation in proton pencil-beam scanning (PBS) can be distinguished based on PGI information using a slit camera [1]. Material and Methods For a virtual head-and-neck tumor in an anthropomorphic head phantom, a PBS treatment plan with simultaneous integrated boost (3 beams, 70Gy and 57Gy in 33 fractions) was generated. For all PBS spots in the investigated beam, PGI profiles were simulated using a verified analytical model of the slit camera [2, 3] for the reference scenario as well as for different error scenarios: SPR change of ±1.0, ±2.0 and ±3.5%, setup error in beam direction of ±1mm and ±3mm, and 10 scenarios of realistic anatomical changes (Fig. 1). A decision-tree approach was proposed to classify different groups of error sources. This included preceding filtering of PBS spots containing reliable PGI information for range verification. For simplification and better hypothesis generation, the head phantom was first overridden with water density. Afterwards, the real phantom anatomy including all heterogeneities was analyzed. It was evaluated whether the different error scenarios could be classified correctly. Results An automated filter to identify reliable PBS spots was developed, e.g. assuring that the spot position is within the effective field of view (FOV) of the camera and that the fall-off of the PGI profile is completely included in the FOV – even in case of range shifts. For subsequent decision-tree-based error source classification (Fig. 2), the following parameters were selected: The coefficient of determination (R2), the slope and intercept of the linear regression between range shift and penetration depth as well as the 2D range shift map. With this approach, 27 of 30 error scenarios could be identified correctly. However, the three error scenarios with anatomical changes in the nasal cavity could not be identified because the automated filtering approach had removed most relevant spots in this region. Conclusion An automated classification approach was introduced to identify the source for range deviation solely from prompt-gamma information. Based on phantom data, including simulation of realistic anatomical variation, the results are promising. Further refinement of this initial approach might be beneficial. An extension of the validation with patient CT data is in preparation. In the future, an application of the approach on clinically measured PGI data is planned. Also other classification methods could be evaluated.

Introduction Currently, the uncertainty in CT-based range prediction is substantially impairing the accuracy of particle therapy. Improvements like a determination of stopping-power ratio (SPR) from dual-energy CT (DECT) have been proposed. However, a validation of range prediction in patients has not been achieved by any means. Here, we present the first verification of proton range prediction in patients, using Prompt-Gamma-Imaging (PGI). Materials & Methods A PGI-slit-camera system was modified to enhance its positioning accuracy using a floor-based docking station. Its accuracy and positioning reproducibility were determined with x-ray and PGI measurements. The PGI system was clinically applied to monitor absolute proton ranges for a 1.5Gy field during eight fractions of hypo-fractionated treatment of two prostate-cancer patients using pencil beam scanning (Fig.1). For all monitored fractions, in-room control-CT scans were acquired in treatment position enabling PGI-based spot-by-spot range analysis for the actual patient anatomy. The PG measurements were compared to simulations of the expected PGI signal using either a standard CT-number-to-SPR conversion (HLUT) or a HLUT optimized with patient-specific DECT-derived SPR information (DECT-HLUT), respectively. Results The accuracy and precision for global range verification (averaging over multiple spots) was determined to be 0.6mm and 1.3mm (both 2σ-level), respectively. The precision is limited by remaining uncertainties in image registration and positioning reproducibility (1mm,2σ). To verify range prediction in patients, the histogram of range shifts was analyzed concerning its Gaussian mean (Fig. 2) as surrogate for the accuracy of the respective range prediction method, independent from random uncertainty contributions (e.g. positioning, statistical uncertainty in shift determination). The mean deviation for the DECT-HLUT and standard HLUT were -0.6mm and 1.3mm, respectively. Conclusion The accuracy of PGI-based range verification was improved to enable the verification of CT-based stopping-power prediction in patients for the first time. First data evaluation suggests a slight superiority of DECT-based range prediction.

Fiberoptical dosimetry uses compact solid state radioluminescence probes coupled to long to flexible light guides. Such probes are convenient, robust small. Especially in the presence of magnetic fields, such optical probes can be advantageous over the transmission of a current signal, e.g. from a photodiode directly attached to a scintillator. These characteristics make such probes attractive for emerging medical applications like particle therapy in combination with MRI. Challenging remains the discrimination of so called “stem” effect light which is generated in the fiber from the actual signal light from the probe. In this work, fiber probes attached to a single photon sensor have been exposed to a proton beam from a proton cyclotron in order to study the feasibility of fiber dosimetry in therapeutic proton fields. Probes with different luminophores have been placed in the beam within a PMMA holder of 1 cm thickness. The light guide was coupled to a Hamamatsu H12386-210 single photon detector. For timing information, the pulse was sampled and analyzed with a Serious Dynamics DAQ125 board. This is a 16 bit sampling ADC board which was running synchronous to the cyclotron. An interpolated time stamp with a resolution of 30 ps was calculated in realtime. It could be shown that a time resolved measurement of the single luminescence photons exhibits the time structure of the luminophore, e.g. the long decay which appears uncorrelated of beryllium oxide or lithium tetraborate or the decay in the ns range of plastic scintillators. Blank fiber measurements exhibit the microbunch width of the accelerator. Thereby, stem identification in therapeutic hadron fields is possible without further reference measurements. Additionally, the issue of quenching of many luminophores in hadron fields can be addressed. A further beamtime at the AGOR cyclotron in Groningen will be conducted in May 2019 and recent results will be presented at the conference.

Introduction Proton therapy is the most advanced radiotherapeutic technique. For the same dose in the tumor, the dose in the surrounding normal tissue is substantially lower than in classical radiation therapy using photon beams. Protons stop in the patient and no dose is deposited behind that position. However, the resulting dose deposition is more sensitive to anatomical changes during treatment and to uncertainties in the CT-based range calculation. Therefore, monitoring of the proton range during treatment would substantially improve proton therapy. Prompt gamma radiation, which is emitted in nuclear reactions of the proton beam with the patient’s tissue, can be used to conclude on the proton range. The most translational advanced approach is prompt-gamma imaging (PGI) using a so-called slit camera [1]. A one-dimensional projection of the prompt-gamma distribution is acquired through a slit collimator onto a spatially resolved detector (Fig.1). The system, developed by IBA, is evaluated concerning its clinical benefit. Materials and Methods The PGI slit-camera was first enabled for routine clinical use by developing dedicated calibration and QA procedures as well as a clinical workflow and setting up a clinical study. In 2015, we were able to perform the first in-man PGI-based range verification [2], 13 years after the idea was initially proposed. In addition, the sensitivity and accuracy of the PGI slit-camera to detect different types of treatment deviations were evaluated for pencil-beam-scanning (PBS) proton therapy in an anthropomorphic phantom [3]. Following, a 2nd generation system was developed to improve positioning accuracy and reproducibility. Clinical application of this system in PBS proton therapy started in 2018. Results In the first in-man applications, detected inter-fractional global range variations were below ±2 mm, and thereby within the overall measurement uncertainties. This was verified with independent control CTs acquired directly before treatment. The phantom study revealed, that global and local range shifts can be detected with high sensitivity. Shift detection accuracy was better than 2 mm under clinical conditions. With the 2nd generation system, positioning uncertainty could be reduced to about 1 mm, potentially allowing a first in-man validation of the absolute range prediction as the measurement uncertainty is below the range prediction uncertainty. Conclusion With the translation from basic physics experiments into clinical operation, we were able to demonstrate the applicability of PGI-based range verification for improving the accuracy of proton therapy. Currently, we systematically evaluate the clinical benefit of the approach aiming for an automated detection of clinical relevant treatment deviation and classification of the underlying error source.

Range verification during clinical treatments is a key for further improving the precision and for reducing the normal-tissue toxicity of radiotherapy with proton beams. In spite of the breakthrough achieved with IBA’s knife-edge slit camera, capable of imaging single beam spots in pencil-beam scanning (PBS) treatment fractions delivered to cancer patients, there are ongoing activities aiming at systems distinguished by lower expense, lower weight, easier integration in the therapy facility, and potentially higher precision. In this context, OncoRay’s Prompt Gamma-Ray Timing (PGT) technology has been further explored with the focus on quantifying uncertainty contributions in potential clinical applications. Experimental data acquired in the treatment room of the University Proton Therapy Dresden (UPTD) during single pencil beam delivery to a polymethyl methacrylate (PMMA) target with arbitrary air cavities have been carefully analyzed. Besides the limited number of events the instable phase relation between proton-beam bunches and accelerating RF turned out to be the weakest point at the given facility. Technical means for monitoring this phase relation at a time scale of split seconds without touching the medical beamline are being developed and will be discussed in the paper. Altogether PGT could provide range verification with 3 mm accuracy at PBS spot level, if (at least) eight PGT detection units are deployed. Technically even a larger number of detectors could be arranged around the nozzle, which would further reduce the uncertainty.

With particle therapy, more and more patients around the world are benefiting from precise dose deposition in the tumor. Due to the characteristic depth dose distribution, however, hadron therapy is particularly susceptible to range inaccuracies. Particle range verification is the subject of current research, but not yet a clinical standard. To circumvent this problem, safety margins are currently being defined around the tumor volume, which nullify the potential precision of particle compared to conventional photon therapy. The use of the prompt gamma radiation resulting from the deceleration of hadrons in tissue for range verification is a promising approach here. At present, various methods exist (for example, prompt gamma-ray imaging, prompt gamma-ray spectroscopy, prompt gamma-ray timing, prompt gamma-ray peak integration), which attempt to obtain information regarding the range from the temporal and / or spatial distribution of these high-energy photons. However, all methods are based directly or indirectly on the results of particle transport calculations. But their results show significant discrepancies compared to the experimental data [1] - [7]. Photon production cross sections are particularly important for range verification with prompt gamma radiation, although there is hardly any experimental data for the clinically relevant isotopes to check and optimize the underlying models. The importance of prompt photon yields in clinical research was therefore also the subject of the 2nd ESTRO Physics Workshop Science and Development "Improving Range Accuracy in Particle Therapy" and will soon be emphasized again in a position paper of the society. At the University Proton Therapy Dresden, the prompt emission spectrum of homogeneous graphite targets of different thickness was determined by irradiation with 90, 150 and 226 MeV protons. The detector response of the CeBr3 scintillation detectors (placed below 55 °, 90 ° and 125 ° with respect to the beam axis) was determined by Geant 4 simulations and verified by measurements with radioactive emitters. The emission spectrum was then obtained by unfolding the detector response using two different deconvolution algorithms (gold deconvolution and spectrum stripping). Scattered protons, which were detected in a YAP / BGO-Phoswich detector below 35°, were used to determine the incident proton fluence. The yields thus obtained are in good agreement with the available experimental data. Bibliography [1] J. Berthold, Single Plane Compton Imaging for Range Verification in Proton Therapy - A Proof-of-Principle Study, Dresden: Technische Universität Dresden, 2018. [2] L. Kelleter, A. Wronska, J. Besuglow, A. Konefał, K. Laihem, J. Leidner und A. Magiera, „Spectroscopic study of prompt-gamma emission for range verification in proton therapy,“ Physica Med., Bd. 34, pp. 7-17, 2017. [3] M. Pinto, D. Dauvergne, N. Freud, J. Krimmer, J. Létang und E. Testa, „Assessment of Geant4 Prompt-Gamma Emission Yields in the Context of Proton Therapy Monitoring,“ Frontiers in Oncology, Bd. 6, 2016. [4] J. Jeyasugiththan und S. Peterson, „Evaluation of proton inelastic reaction models in Geant4 for prompt gamma production during proton radiotherapy,“ Phys. Med. Biol., Bd. 60, p. 7617–7635, 2015. [5] A. Schumann, J. Petzoldt, P. Dendooven, W. Enghardt, C. Golnik, F. Hueso-González, T. Kormoll, G. Pausch, K. Roemer und F. Fiedler, „Simulation and experimental verification of prompt gamma-ray emissions during proton irradiation,“ Phys. Med. Biol., Bd. 60, pp. 4197-4207, 2015. [6] J. Verburg, Reducing Range Uncertainty In Proton Thearpy, Eindhoven: Technische Universiteit Eindhoven, 2015. [7] J. Dudouet, D. Cussol, D. Durand und M. Labalme, „Benchmarking Geant4 nuclear models for hadron therapy with 95 MeV/nucleon carbon ions,“ Phys. Rev. C, Bd. 89, p. 054616, 2014.

Introduction The accuracy of proton therapy is basically delimited by range uncertainties. In order to fully exploit the benefit of proton therapy, real-time range verification is needed. Prompt gamma rays, produced during the deceleration of protons in tissue, allow assessing the proton range indirectly. Prompt gamma-ray timing (PGT) utilizes measured time distributions of prompt gamma rays for estimating the beam range [1-4]. At the given therapy facility, however, the accuracy of range determination is affected by drifts of the proton bunch timing (or phase) with respect to the cyclotron radiofrequency, and by the proton bunch width. To counter this limitation, a proton bunch monitor (PBM) has to be introduced. Materials & Methods The coincident detection of protons elastically scattered out of a thin target allows measuring phase and width of the proton micro-bunches [3]. Two CeBr3 scintillation detectors coupled with PGT electronics [4] were employed for these measurements. Alternatively, scintillating fibers placed in the beam halo have also been studied. Results The applicability of standard PGT detectors and electronics [4] for accurate bunch phase and width measurements has been demonstrated. The detection of beam-halo protons with scintillating fibers seems to be a feasible alternative for bunch phase monitoring at higher sampling rates. Quantitative analysis are still ongoing. Summary Alternative proton bunch monitoring approaches have been investigated to assess beam parameters necessary for range verification by means of the PGT method. An appropriate PBM would complete the current PGT detection system. Acknowledgement This work has been supported by StrahlenSchutzSeminar in Thüringen e.V. Literature [1] C. Golnik et al., Phys. Med. Biol. 59 (2014) 5399 [2] F. Hueso-González et al., Phys. Med. Biol. 60 (2015) 6247 [3] J. Petzoldt et al., Phys. Med. Biol. 61 (2016) 2432 [4] T. Werner et al., Phys. Med. Biol. accepted (2019)

Anger cameras are still the primary technology for radionuclide imaging in nuclear medicine [1]. Despite the achieved advances in improving the image quality over the past 60 years since its invention [2], there are physical limits to the camera performance (limited detection efficiency, decreasing spatial resolution of high energetic gamma rays, fixed dependency of the spatial resolution and detection efficiency from the used collimator). In order to overcome these limitations, the concept of the “Single Plane Compton Camera” (SPCC, see also Ref. [3]) was developed. The SPCC is based on the idea of the “Directional Gamma Radiation Detector” published in Ref. [4] and [5]. A setup for the investigation of the SPCC concept was developed recently at the Helmholtz-Zentrum Dresden - Rossendorf. Based on a GAGG:Ce (Gadolinium Aluminum Gallium Garnet, Gd3Al2Ga3O12) scintillator array and read out by digital silicon photomultipliers the setup is intend to deliver spatial information of activity distributions. The authors will present first experimental results acquired with the new setup and will compare them to predictions obtained from particle transport calculations performed with GEANT 4 [6]. [1] S. R. Cherry, J. A. Sorenson and M. E. Phelps, Physics in Nuclear Medicine, 4th ed., Elsevier, 2013. [2] H. Anger, "A new instrument for mapping gamma ray emitters," Biology and Medicine Quaterly Report, 1957. [3] G. Pausch et al., Paper N60-1 presented at the 2016 IEEE NSS/MIC in Strasbourg, France, Conference Record [4] G. Kraft et al., U.S. patent no. 8030617 B2, granted in Oct. 2011 [5] A. Gueorguiev et al. U.S. patent no. 8299441 B2, granted in Oct. 2012 [6] A. Allison, "Geant 4 - a simulation toolkit", Nucl. Instr. and Meth. A. 506, (2003) 250-303

The paper reports on first attempts to prove the concept of Single Plane Compton Imaging (SPCI), recently proposed on the basis of simulations, in practice. SPCI combines electronic collimation as known from conventional Compton cameras with a much simpler detector design: Multiple scintillator pixels are arranged alongside in a single detection plane. Imaging information is encoded in a set of ‘conditional’ spectra meaning energy deposition distributions in single pixels obliged with the condition of a coincident detection in another (adjacent) pixel. The activity distribution is iteratively reconstructed from the measured projections (the bin contents of the conditional spectra) by using Maximum Likelihood Expectation Maximization (MLEM) algorithms. This concept has been approached experimentally with two distinct setups addressing the application fields of radionuclide imaging in nuclear medicine, and of prompt-gamma based range verification in radiooncology with proton beams. The first setup consists of a 4×4 array of about 7 × 7 × 20 mm3 GAGG scintillator pixels read out with a Philips STEK module comprising 4×4 digital silicon photomultiplier dies. Data were taken with radioactive point sources arranged in few-cm distance from the scintillator pixels. The second setup consists of much larger monolithic cerium bromide scintillation detectors arranged head-to-head in pairs. Those were exposed to prompt gamma radiation produced by a 90 MeV proton beam in a beam-stopping polymethyl acrylate (PMMA) target. Though data analyses are not yet finished, the effects enabling imaging are clearly visible. Preliminary plots exemplify the applicability of SPCI in both applications. The experimental activities have been closely accompanied with appropriate modeling using the Geant4 toolkit

Radionuclide imaging is a well-established and important diagnostic tool. The collected data contain physiological information as opposed to other traditional anatomical imaging techniques like X-ray computed tomography. The Anger Camera remains the primary device for nuclear medical imaging since its invention in 1957 [1]. However, despite the achieved advance in the imaging quality through the years, there are still performance limitations, resulting from the limited detection efficiency, reduced spatial resolution of highly energetic gamma rays, fixed dependency of the spatial resolution and detection efficiency from the used collimator and others. In order to overcome these limitations, the concept of “Single Plane Compton Imaging” (SPCI) has been proposed [2]. This concept, based on the idea of a “Directional Gamma Radiation Detector” [3] - [4] is now being explored experimentally. We will present first experimental results obtained with a GAGG scintillator pixel array read out by a digital silicon photomultiplier array. The detector is operated in our laboratory and exposed to gamma radiation emitted from radioactive sources. Information about the source positions is derived from the energy spectra of coincident events in adjacent scintillator pixels.

CeBr3 scintillation detectors with light readout by photomultiplier tubes (PMT) have been used for prompt-gamma measurements in proton radiotherapy aiming at treatment verification by Prompt Gamma Timing (PGT) or Prompt Gamma Spectroscopy (PGS). Such treatments are usually structured in distinct beam spots grouped in several mono-energy layers, separated by breaks of few-seconds duration for beam energy switching. This causes a multitude of extreme load steps during delivery of a single treatment field. The paper presents preliminary results of two experiments exploring effects of such load steps on gain and timing of PGT detection units as developed for use in clinical treatments. Multiple units, consisting of scintillation detector (2”×1…2” CeBr3 crystal coupled to Hamamatsu R13089 or R13089-100 photomultiplier) and a plugged-on high-throughput digital spectrometer (U100 by Target Systemelektronik) that also provides well stabilized dynode voltages for the PMT, were exposed to prompt gamma radiation produced by 225 MeV protons in a plexiglass (PMMA) layer, or to bremsstrahlung produced by 13 MeV electrons. Beam shots of 3-5 s duration and varied intensity provoked load steps from background up to the Mcps region and back, and allowed analyzing the immediate and the retarded response of PMT gain and timing. Indeed we observed a noticeable change of the PMT transit time with the detector load, indicating that space charge effects are involved. The scaling of gain turns with the mean anode current supports this hypothesis. As long as the mean anode current is in a ‘reasonable’ operating range, gain and timing drifts of given detectors are well correlated, at least in the stationary case. The observed load-induced timing shifts are as large as 100 ps and would seriously disturb PGT measurements in a treatment, but could eventually be corrected for by tracking the PMT gain. In treatments, this could be done by tracking the ubiquitous 511 keV annihilation peak.

Positron emission tomography (PET) is a means of imaging the β+-activity produced by the radiation field in ion beam therapy and therefore for treatment verification. Prompt Γ-rays that are emitted during beam application challenge the detectors and electronics of PET systems, since those are designed for low and medium count rates. Typical PET detectors operated according a modified Anger principle suffer from multiple events at high rates. Therefore, in-beam PET systems using such detectors rely on a synchronization of beam status and measurement to reject deteriorated data. In this work, a method for pile-up rejection is applied to conventional Anger logic block detectors. It allows for an in-beam data acquisition without further synchronization. Though cyclotrons produce a continuous wave beam, the radiation field shaping technique introduces breaks in the application. Time regimes mimicking synchrotrons as well as cyclotron based ones using double-scattering or pencil beam scanning field shaping at dose rates of 0.5, 1.0 and 2.0 Gy/min were investigated. Two types of inhomogeneous phantoms were imaged. The first one simulates cavity structures, the other one mimics a static lung irradiation. It could be shown that, depending on the dose rate and the beam time structure, in-beam measurement including a few seconds decay time only, yield images which revealed all inhomogeneities in the phantoms. This technique can be the basis for the development of an in-beam PET system with traditional detectors and off-the-shelf electronics.

Background and Purpose: Prompt-gamma imaging (PGI) was recently applied successfully in first clinical patient treatments in pencil beam scanning (PBS) and double scattering (DS). Still, systematic evaluations on its capability in clinical conditions are desirable. Here, the performance of the slit-camera is systematically assessed in well-defined error scenarios using realistic treatment deliveries to an anthropomorphic phantom. Materials and Methods: The sensitivity to detect global and local range shifts with the slit camera was investigated in PBS and DS irradiations of a head phantom. For PBS, measured PGI information for shifted geometries was compared spot-wise with either simulated or measured un-shifted PGI-information to evaluate the sensitivity to detect deviations from the treatment plan and interfractional shifts, respectively. Results: Deviations from the treatment plan can be detected with an accuracy of 1.5 and 3 mm for global and local shifts in PBS, respectively. Interfractional comparisons are more affected by noise in the measurements. Evaluation of the average PGI signal of the whole field allows the detection of global shifts also in DS mode. Conclusions: PGI-based detection of global and local range shifts under clinical conditions is possible. Especially for PBS treatments, both high sensitivity and high accuracy in shift detection were found.

To ensure the optimal outcome of proton therapy, in-vivo range verification is highly desired. Prompt gamma-ray imaging (PGI) is a possible approach for in-vivo range monitoring. For PGI dedicated detection systems, e.g. Compton cameras, are currently under investigation. The presented paper deals with substantial requirements regarding hardware and software that a Compton camera used in clinical routine has to meet. By means of GEANT4 simulations, we investigate the load on the detectors and the percentage of background expected in a realistic irradiation and we simulate gamma-ray detections, i.e. input data for the reconstruction. By reconstructing events from simulated sources of well-defined geometry, we show that large-area detectors are favourable. We determine the minimum number of valid events allowing for a statistically significant range assessment. Finally, an end-to-end test for a realistic patient scenario is presented: starting with a treatment plan, the gamma-ray emissions are calculated, the detector response is modelled, and the image reconstruction is performed. We conclude that, with respect to the achievable precision, the expected complexity, and high costs of an adequate detection system, in-vivo dosimetry with Compton cameras will hardly be realised.

Purpose The dose distribution and range of proton beams are exceedingly prone to uncertainties and anatomical changes, demanding for an in-vivo range verification. A promising approach is prompt γ-ray imaging (PGI), which was recently implemented clinically in Dresden using a so-called PGI slit camera [1,2] in double scattering (DS). However, the detectability of local range shifts, affecting only part of the lateral field in DS, is limited. The spot-wise dose deposition in pencil beam scanning (PBS) promises a finer spatial resolution of range shifts. The purpose of this study is to comprehensively investigate the sensitivity to detect range shifts in DS and PBS using a head phantom in a clinical setup. Materials and Methods For a realistic brain tumor treatment, treatment plans in DS and PBS (2 beams, 60 GyE, 2 GyE/fx), were created. One beam (1 GyE) was applied to a CIRS head phantom and monitored with the PGI slit camera. To investigate the influence of the spot dose, the same beam with 5 GyE was also delivered and measured. Global and local (5 cm in diameter) range shifts were introduced and the PGI profiles (prompt-γ counts over depth) with and without shifts were compared. Sum profiles containing prompt-γ counts over the entire fraction were used for the comparison of DS and PBS. Moreover, PGI profiles measured in PBS were analyzed spot-wise and will also be compared with simulated profiles for absolute range determination. Results A good agreement between introduced and measured global shifts was found in the sum profile evaluation for both modalities, PBS and DS (Table 1). Relative differences were below 2, 7 and 12 % for the 10, 7 and 4 mm shifts, respectively. Local shifts are not detectable using sum profiles. For the applied local shifts, a spot-wise comparison of PGI profiles in PBS allows the detection and localization of global and local shifts (Figure 1). For interpretation, neighboured spots should be clustered, as shifts detected for single spots are less reliable due to low statistics. Higher doses (5 vs. 1 GyE) allow the detection of smaller shifts as shown in Figure 1 for the 4 mm local shift. Conclusion The systematic sensitivity study revealed the capability of the PGI slit camera to detect range shifts under clinical conditions. In both treatment modalities, global range shifts can be detected. Additionally, in PBS a spot-wise comparison allows also the determination of interfractional local range shifts. Moreover, a still ongoing evaluation of PBS measured and simulated spot-wise profiles for absolute range verification will be presented.

Proton beams are promising means for treating tumors. Such charged particles stop at a defined depth, where the ionization density is maximum. As the dose deposit beyond this distal edge is very low, proton therapy minimizes the damage to normal tissue compared to photon therapy. Nevertheless, inherent range uncertainties cast doubts on the irradiation of tumors close to organs at risk and lead to the application of conservative safety margins. This constrains significantly the potential benefits of protons over photons. In this context, several research groups are developing experimental tools for range verification based on the detection of prompt gammas, a nuclear by-product of the proton irradiation. At OncoRay and Helmholtz-Zentrum Dresden-Rossendorf, detector components have been characterized in realistic radiation environments as a step toward a clinical Compton camera. On the one hand, corresponding experimental methods and results obtained during the ENTERVISION training network are reviewed. On the other hand, a novel method based on timing spectroscopy has been proposed as an alternative to collimated imaging systems. The first tests of the timing method at a clinical proton accelerator are summarized, its applicability in a clinical environment for challenging the current safety margins is assessed, and the factors limiting its precision are discussed.