Beam characteristics of the first clinical 360° rotational single gantry room scanning pencil beam proton treatment system and comparisons against a multi‐room system

Abstract Purpose The purpose of this study was to present the proton beam characteristics of the first clinical single‐room ProBeam Compact™ proton therapy system (SRPT) and comparison against multi‐room ProBeam™ system (MRPT). Materials and Methods A newly designed SRPT with proton beam energies ranging from 70 to 220 MeV was commissioned in late 2019. Integrated depth doses (IDDs) were scanned using 81.6 mm diameter Bragg peak chambers and normalized by outputs at 15 mm WET and 1.1 RBE offset, following the methodology of TRS 398. The in‐air beam spot profiles were acquired by a planar scintillation device, respectively, at ISO, upper and down streams, fitted with single Gaussian distribution for beam modeling in Eclipse v15.6. The field size effect was adjusted for the best overall accuracy of clinically relevant field QAs. The halo effects at near surface were quantified by a pinpoint ionization chamber. Its major dosimetric characteristics were compared against MRPT comparable beam dataset. Results Contrast to MRPT, an increased proton straggling in the Bragg peak region was found with widened beam distal falloffs and elevated proximal transmission dose values. Integrated depth doses showed 0.105–0.221 MeV (energy sigma) or 0.30–0.94 mm broader Bragg peak widths (Rb80–Ra80) for 130 MeV or higher energy beams and up to 0.48–0.79 mm extended distal falloffs (Rb20–Rb80). Minor differences were identified in beam spot sizes, spot divergences, proton particles/MU, and field size output effects. High passing scores are reported for independent end‐to‐end dosimetry checks by IROC and for initial 108 field‐specific QAs at 3%/3 mm Gamma index with fields regardless with or without range shifters. Conclusions The author highlighted the dosimetry differences in IDDs mainly caused by the shortened beam transport system of SRPT, for which new acceptance criteria were adapted. This report offers a unique reference for future commissioning, beam modeling, planning, and analysis of QA and clinical studies.


| INTRODUCTION
The first 360°rotational single gantry room scanning pencil beam proton treatment system (SRPT) -ProBeam Compact™ (Varian Medical, Palo Alto, CA) was implemented in a clinical setting in November 2019. This system mainly consists of (a) a superconducting cyclotron which accelerates and injects 250 MeV protons to the beamline; (b) open-air energy selection system 1 with carbon multiwedge technique for clinical beam energies ranging from 70 to 220 MeV; (c) shortened single-room dedicated beam transport system removing two or three entrance bending magnets keeping a 45°a nd a 135°major bending magnets; 2 (d) 360°rotating gantry equipped by two orthogonally arranged onboard kV imaging systems with CBCT capability; and (e) six-dimensional (6D) robotic patient support system. Different from a conventional multi-gantry ProBeam proton treatment system (MRPT), the newly designed components two and three contribute unique dosimetric characteristics of the scanning proton beams and there is no report of clinically relevant beam parameters, we attempt to fill this void here.
The principle beam dosimetric components of the commissioning typically comprised of integrated depth dose curves (IDDs), absolute dose calibration for given MUs (or dose output), in-air beam spot size or profiles. [3][4][5][6] Other essential elements are virtual source position relative to ISO, beam spot accuracy and dose uniformity, field size factors (or halo effect of spot profile), MU linearity, mechanical accuracy, OBI and CBCT quality and accuracy, CT stoichiometric calibration, WET measurement for the range shifters, table support and inserts, as well as immobilization and physics accessories. [7][8][9] However, only beam dosimetric characteristics will be discussed in this report.

| MATERIALS AND METHODS
The commissioning of this ProBeam Compact™ was conducted in November 2019. IDD data were acquired by PTW 81.6 mm diameter Bragg peak plane-parallel chambers (Model 34070 primary and model 34080 reference) and a PTW 3D Water Scanning System MP30PL (Freiburg, Germany), using central axial downward (AP) proton beams in every 5 MeV energy intervals. Following the methodology recommended by TRS 398 report, the absolute dose output of each nominal monogenic beam was obtained at 15 mm depth in water aligned to the ISO using an ADCL calibrated PPC05 Markus parallel-plate chamber (IBA Dosimetry). In addition, the known dosimetric output accuracy issues in PCS 3,4 were corrected with Acur-osPT calculations for different test fields. 7 To convert the measurements in the transmission beam region to radiation doses, 1.002 k Q factor was used along with 1.1 RBE offset factor, which were then imported to an Eclipse v15.6 (Varian, Palo Alto, CA) for modeling the PCS and NUPO algorithms. Here, the low-energy (85 to 70 MeV) dose outputs were slightly adjusted by using 0.995-0.990 for k Q variations. 10 The final IDD outputs particularly for AcurosPT were fine-tuned for an optimal overall accuracy, based on the measurements of the calibrated ionization chamber two-dimensional (2D) array on different testing field sizes and various patient plan-specific QAs. 3,4 In this commissioning, AcurosPT algorithm was exclusively utilized for final planning dose computations.
The measurements of in-air proton beam spots were accom-   The analysis of patient-specific QAs for over 108 proton fields, measured by Octavius-1500 XDR ionization chamber 2D array, is summarized in Table 3, the average passing rates exceeded 95% and all comparisons pass 90% using Gamma index criteria of 3 mm, 3% (local), for fields with or without range shifters. All beams were  vendor. 19 In comparing with the scanning proton beam data from a recent multi-room ProBeam, SRPT exhibited a widened range straggling in the Bragg peaks and elevated proximal transmission beam dose, especially for the proton energies of 130 MeV or greater. This suggests that of the energy spectrum of the delivered proton beams for a given nominal proton energy is less confined, due to the challenges from the truncated beam transport system with inability of stripping off all the outlier energies from the energy selection assembly. 2 When 250 MeV protons from the cyclotron pass through its energy selection system, the interactions of protons with the multiwedged attenuator create a narrow range of energy spectrum around the selected MeV which, after passing through a beam aperture, is sent to the beam transportation line on the gantry after. The energy spectrum of the proton beam will be further narrowed in the remain- While the surface dose discrepancies increase more drastically with lower energy proton beams (Fig. 4), the majority of shallow dose is contributed by more consistent transmission doses of higher energy protons. Thus, in most of cases, the surface dose deviation could still be within the clinically acceptable range. Employment of a range shifter when applicable to avoid lowest energies can also further minimize the surface dose deviations for a shallow small target. Table 4, the surface dose disagreement for small T A B L E 1 Near surface (WET = 8 mm) dose deviations between the computed by Eclipse AcurosPT and the measured by PTW Octavius ion chamber array andthe doses acquired by PTW Semilex 0.07 cc ion chamber using 70 MeV single-layer proton beams at different field sizes.

ACKNOWLEDGMENTS
Authors would like to thank Varian engineering team, Proton International LLC for providing unparalleled technical and logistic support; Tim Williams, MD for guiding on clinical implementation.

CONF LICT OF I NTEREST
None.