A proton range telescope for quality assurance in hadrontherapy
2009 IEEE Nuclear Science Symposium Conference Record (NSS/MIC)
Executive summary Hadrontherapy is currently a clinical reality in radiation oncology and a proven technique in the fight against cancer. In the world today, hadrontherapy is being more and more widely employed for treating patients with non-operable deep-seated or radio-resistant tumours because of its advantage in delivering a highly conformal dose to the tumour volume. This offers an increased likelihood of tumour control and a better sparing of healthy surrounding tissue as compared with
... ditional radiotherapy which use photon beams. Despite the fact that 35 centers are currently treating patients, hadrontherapy is still considered to be an emerging clinical technique. One of the persisting challenges to hadrontherapy is the verification of the dose delivered to the patient since the physical properties of hadrons are only beneficial for therapy if they can be delivered precisely to the tumour volume. Quality assurance can be achieved using novel diagnostic techniques which make use of radiation detectors similar to those developed for highenergy physics experiments and already used in medical imaging. Proton radiography can be used to verify the patient setup prior to irradiation, using a diagnostic proton beam of higher energy and lower intensity, but can also provide directly the information needed for accurately computing the range of hadrons in the patient tissues. Range calculations currently rely on X-ray CT data, and are characterized by a small but non-negligible uncertainty. During irradiation with the therapeutic beam, the activation of the patient tissues caused by nuclear interactions with the hadron beam can be visualized by PET detectors, making it possible to perform in-vivo dosimetry during irradiation and in the minutes immediately following. In this context, this thesis presents an expansive study of novel radiation detectors which have been developed for quality assurance in clinical hadrontherapy. Three distinct detector solutions are described, a proton radiography instrument and two detectors technologies which could be used for performing in-vivo dosimetry of the delivered treatment plan. In the case of proton range radiography (PRR), a novel instrument called the PRR10 has been built having 10x10 cm 2 active area and covering a residual range of 10 cm water-equivalent path length (WEPL). The PRR10 has been extensively tested with proton beams at the Paul Scherrer Institute (PSI) in Villagen, Switzerland and at the Centro Nazionale di Adroterapia Oncologica (CNAO) in Pavia, Italy. A residual range resolution of 1.6 mm WEPL has been measured as well as a spatial resolution better than 1 mm. The PRR10 currently sits at the CNAO center awaiting further testing while a new instrument, the PRR30, which has an active area of 30x30 cm 2 , is reaching a final stage of completion. The PRR30 will allow full-size PRR images to be made and is scheduled for testing with proton beams at PSI and the CNAO by the end of 2013. 3 To perform in-vivo dosimetry, two different PET technologies have been studied. The first is based on inorganic scintillators (crystals) coupled to a photodetector, having many similarities to conventional PET hardware for nuclear medicine. The design for a unit PET detector based on crystal follows the trends in current PET research allowing for the depth-of-interaction (DOI) to be measured as well as the time-of-flight (TOF) between the coincidence photons. Both techniques result in a higher effective sensitivity and a better rejection of noise, and therefore higher quality PET images. Two prototypes have been assembled and tested, built using 12x60x30 cm 3 LYSO crystals and a multi-anode Multi-Channel Plate (MCP) photodetector. An excellent localization of the photon interaction, 1.2 mm in the transverse direction and ∼15 mm in DOI, have been demonstrated with an energy resolution of 13% FWHM. The coincidence TOF resolution has been measured as 810 ps. The second PET technology we have studied makes use of multi-gap Resistive Plate Chambers (MRPCs), which are highly unusual in PET because of their low detection efficiency to 511 keV gammas. Compact MRPC modules have been built and tested, having 7x10 cm 2 and 12x30 cm 2 active area. The design and assembly procedure has been shown to be suitable for mass-production, a requirement for overcoming the intrinsic low efficiency. A 4-gap 7x10 cm 2 MRPC module has been tested and shown to have an efficiency of (0.66±0.01)% to 511 keV gammas. In addition, the timing between ends of the strip readout at either side of the module has been measured as 3.8 ps, enough to allow an interaction localization of 3.5 mm. The single-detector TOF resolution between two single-gap RPCs and two 4-gap MRPCs has been measured as 310 ps and 370 ps, respectively with a coincidence resolution of ∼150 ps expected shortly. To compliment the experimental results, Monte-Carlo simulations of both LYSO-MCP and MRPC-based PET scanners have been carried out using the GATE toolkit. Two commercial detectors, the Philips Gemini and Siemens HiRez, have also been included in the study as a benchmark for the results. The full-ring LYSO-MCP scanner has been shown to have a 57% higher sensitivity than the Gemini to a 70 cm long line source, a consequence of the increased depth (30 mm) of the LYSO crystals used in our design. An MRPC-PET scanner, after performing a sensitivity optimization of various parameters, has been shown to be a factor of 2.5 higher than the Gemini. Although considerable development will be required to build such a MRPC-PET scanner, the gains in sensitivity over existing commercial scanners, coupled with their excellent TOF resolutions, make this technology an exciting alternative to crystals, whether for hadrontherapy quality assurance, or whole-body PET imaging.