Advertisement

Reduction of intrafraction pancreas motion using an abdominal corset compatible with proton therapy and MRI

  • Sergej Schneider
    Affiliations
    Technische Universität Dresden, Faculty of Medicine Carl Gustav Carus, Dresden, Germany

    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Institute of Radiooncology – OncoRay, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany
    Search for articles by this author
  • Sarah Stefanowicz
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Institute of Radiooncology – OncoRay, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany

    Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University Munich, Munich, Germany
    Search for articles by this author
  • Christina Jentsch
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
    Search for articles by this author
  • Fabian Lohaus
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
    Search for articles by this author
  • Julia Thiele
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
    Search for articles by this author
  • Danilo Haak
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
    Search for articles by this author
  • Chiara Valentini
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
    Search for articles by this author
  • Ivan Platzek
    Affiliations
    Department of Radiology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
    Search for articles by this author
  • Esther G. C. Troost
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Institute of Radiooncology – OncoRay, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany

    German Cancer Consortium (DKTK), Partner Site Dresden, and German Cancer Research Center (DKFZ), Heidelberg, Germany

    National Center for Tumor Diseases (NCT), Partner Site Dresden, Germany: German Cancer Research Center (DKFZ), Heidelberg, Germany; Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany; and Helmholtz Association / Helmholtz-Zentrum Dresden - Rossendorf (HZDR), Dresden, Germany
    Search for articles by this author
  • Aswin L. Hoffmann
    Correspondence
    Corresponding author at: OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf Germany, Fetscherstrasse 74, Dresden 01307, Germany.
    Affiliations
    OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Institute of Radiooncology – OncoRay, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany

    Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
    Search for articles by this author
Open AccessPublished:November 08, 2022DOI:https://doi.org/10.1016/j.ctro.2022.11.006

      Highlights

      • PE based abdominal corset allows compression in MRI and PT.
      • Pancreas motion in patients with abdominal tumours reduced significantly.
      • Patient comfort during compression found high.

      Abstract

      Background and purpose

      Motion mitigation is of crucial importance in particle therapy (PT) of patients with abdominal tumors to ensure high-precision irradiation. Magnetic resonance imaging (MRI) is an excellent modality for target volume delineation and motion estimation of mobile soft-tissue tumors. Thus, the aims of this study were to develop an MRI- and PT-compatible abdominal compression device, to investigate its effect on pancreas motion reduction, and to evaluate patient tolerability and acceptance.

      Materials and methods

      In a prospective clinical study, 16 patients with abdominal tumors received an individualized polyethylene-based abdominal corset. Pancreas motion was analyzed using time- and phase resolved MRI scans (orthogonal 2D-cine and 4D MRI) with and without compression by the corset. The pancreas was manually segmented in each MRI data set and the population-averaged center-of-mass motion in inferior-superior (IS), anterior-posterior (AP) and left–right (LR) directions was determined. A questionnaire was developed to investigate the level of patient acceptance of the corset, which the patients completed after acquisition of the planning computed tomography (CT) and MRI scans.

      Results

      The corset was found to reduce pancreas motion predominantly in IS direction by on average 47 % − 51 % as found in the 2D-cine and 4D MRI data, respectively, while motion in the AP and LR direction was not significantly reduced. Most patients reported no discomfort when wearing the corset.

      Conclusion

      An MRI- and PT-compatible individualized abdominal corset was presented, which substantially reduced breathing-induced pancreas motion and can be safely applied with no additional discomfort for the patients. The corset has been successfully integrated into our in-house clinical workflow for PT of tumors of the upper abdomen.

      Keywords

      Introduction

      The treatment of borderline-resectable or unresectable pancreatic cancer using radiation therapy (RT) with photons or protons is increasingly being offered to patients [
      • Hiroshima Y.
      • Fukumitsu N.
      • Saito T.
      • Numajiri H.
      • Murofushi K.N.
      • Ohnishi K.
      • et al.
      Concurrent chemoradiotherapy using proton beams for unresectable locally advanced pancreatic cancer.
      ,
      • Hong T.S.
      • Ryan D.
      • Borger D.
      • Blaszkowsky L.S.
      • Yeap B.Y.
      • Ancukiewicz M.
      • et al.
      A phase 1/2 and biomarker study of preoperative short course chemoradiation with proton beam therapy and capecitabine followed by early surgery for resectable pancreatic ductal adenocarcinoma.
      ,
      • Jethwa K.R.
      • Tryggestad E.J.
      • Whitaker T.J.
      • Giffey B.T.
      • Kazemba B.D.
      • Neben-Wittch M.A.
      • et al.
      Initial experience with intensity modulated proton therapy for intact, clinically localized pancreas cancer: Clinical implementation, dosimetric analysis, acute treatment-related adverse events, and patient-reported outcomes.
      ,
      • Kim T.H.
      • Lee W.J.
      • Woo S.M.
      • Oh E.S.
      • Youn S.H.
      • Jang H.Y.
      • et al.
      Efficacy and feasibility of proton beam radiotherapy using the simultaneous integrated boost technique for locally advanced pancreatic cancer.
      ,
      • Rutenberg MS
      • Nichols RC
      Proton beam radiotherapy for pancreas cancer.
      ]. However, this tumor type still poses a challenge for RT due to the tumor movement induced by respiration and abdominal organ filling, the radiation sensitivity of surrounding organs-at-risk as well as the poor soft-tissue contrast of computed tomography (CT) imaging, making accurate target volume definition difficult [
      • Reese A.S.
      • Lu W.
      • Regine W.F.
      Utilization of intensity-modulated radiation therapy and image-guided radiation therapy in pancreatic cancer: is it beneficial?.
      ]. For such moving tumors, appropriate immobilization of the target volume is a key component in the treatment process for an accurate and precise dose delivery. This is of particular importance in particle therapy (PT) because of range uncertainties due to the higher sensitivity to density variations compared to conventional photon-based RT. Morphological changes along the beam path due to organ motion, deformation and filling can influence the position of the Bragg peak relative to the target volume. Furthermore, the use of actively scanned dose delivery techniques, which employ pencil beams to volumetrically scan a mono-energetic Bragg peak over the target volume, in combination with intrafraction organ motion may further degrade the precision of dose delivery due to the interplay effect [
      • Bert C.
      • Grözinger S.O.
      • Rietzel E.
      Quantification of interplay effects of scanned particle beams and moving targets.
      ,
      • Phillips M.H.
      • Pedroni E.
      • Blattmann H.
      • Boehringer T.
      • Coray A.
      • Scheib S.
      Effects of respiratory motion on dose uniformity with a charged particle scanning method.
      ]. This may result in over- or underdosage of the target volume and unwanted dose deposition in adjacent organs-at-risk.
      For thoracic and abdominal organs, respiration is the main source of intrafraction motion. To reduce respiration-induced uncertainties in PT, breathing motion can be accounted for by breathing phase-resolved 4D CT based treatment planning and robust dose optimization, rescanning, tumor gating, or in future possibly tracking [
      • Graeff C.
      Motion mitigation in scanned ion beam therapy through 4D-optimization.
      ,
      • Bert C.
      • Durante M.
      Motion in radiotherapy: particle therapy.
      ,
      • Dolde K
      • Dávid C
      • Echner G
      • Floxa R
      • Hentschke C
      • Maier F
      • et al.
      4DMRI-based analysis of inter- and intrafractional pancreas motion and deformation with different immobilization devices.
      ,
      • Knopf A.C.
      • Hong T.S.
      • Lomax A.
      Scanned proton radiotherapy for mobile targets the effectiveness of re-scanning in the context of different treatment planning approaches and for different motion characteristics.
      ,
      • Bert C.
      • Saito N.
      • Schmidt A.
      • Chaudhri N.
      • Scardt D.
      • Rietzel E.
      Target motion tracking with a scanned particle beam.
      ,
      • Mori S.
      • Yanagi T.
      • Hara R.
      • Sharp G.C.
      • Asakura H.
      • Kumagai,
      • et al.
      Comparison of respiratory-gated and respiratory-ungated planning in scattered carbon ion beam treatment of the pancreas using four-dimensional computed tomography.
      ,
      • Yamada T.
      • Takao S.
      • Koyano H.
      • Nihongi H.
      • Fujii Y.
      • Hirayama S.
      • et al.
      Validation of dose distribution for liver tumors treated with real-time-image gated spot-scanning proton therapy by log data based dose reconstruction.
      ]. However, their respective use has shown to not completely reduce motion effects [
      • Kraus K.M.
      • Heath E.
      • Oelfke U.
      Dosimetric consequences of tumour motion due to respiration for a scanned proton beam.
      ,
      • Zhang Y.
      • Huth I.
      • Weber D.C.
      • Lomax A.J.
      A statistical comparison of motion mitigation performances and robustness of various pencil beam scanned proton systems for liver tumour treatments.
      ]. Furthermore, both tumor delineation and margin definition based on 4D CT imaging has shown limited accuracy [
      • Reese A.S.
      • Lu W.
      • Regine W.F.
      Utilization of intensity-modulated radiation therapy and image-guided radiation therapy in pancreatic cancer: is it beneficial?.
      ,
      • Yamazaki H.
      • Nishiyama K.
      • Tanaka E.
      • Koiwai K.
      • Shikama N.
      • Yoshinori I.
      • et al.
      Dummy run for a phase II multi-institute trial of chemoradiotherapy for unresectable pancreatic cancer: inter-observer variance in contour delineation.
      ]. Therefore, particularly when no active motion mitigation techniques such as beam gating or tracking are available, physically reducing respiration-induced motion by abdominal compression techniques becomes crucial to avoid extensive margin expansion of the planning target volume. Moreover, as opposed to CT imaging, magnetic resonance imaging (MRI) offers considerable advantages for soft-tissue delineation and is hence considered an ideal modality for target volume delineation of the pancreas and organs-at-risk, as well as for margin reduction and motion management [
      • Fontana G
      • Riboldi M
      • Gianoli C
      • Chirvase CI
      • Villa G
      • Paganelli Chiara
      • et al.
      MRI quantification of pancreas motion as a function of patient setup for particle therapy —a preliminary study.
      ,
      • Heerkens HD
      • Reerink O
      • Intven MPW
      • Hiensch RR
      • van den Berg CAT
      • Crijns SPM
      • et al.
      Pancreatic tumor motion reduction by use of a custom abdominal corset.
      ,
      • Dolde K.
      • Naumann P.
      • Dávid C.
      • Gnirs R.
      • Kachelrieß M.
      • Lomax A.J.
      • et al.
      4D dose calculation for pencil beam scanning proton therapy of pancreatic cancer using repeated 4DMRI datasets.
      ].
      For PT of targets in the upper gastro-intestinal tract, motion mitigation by means of abdominal compression belts or pressure plates can exacerbate the range uncertainties whenever the devices extend into the treatment fields due to a limited setup reproducibility and consequential edge effects [
      • Wroe A.J.
      • Bush A.D.
      • Slater J.D.
      Immobilization Considerations for Proton Radiation Therapy.
      ]. In particular, pressure plates incorporated into a frame system restrict the usability in MR-guided PT, since their bulky setup can prohibit positioning of the PT beam nozzle close to the patient’s surface, thereby increasing lateral beam spread [
      • Newhauser WD
      • Zhang R
      The physics of proton therapy.
      ]. In addition, their material composition and size may be incompatible with MRI. Moreover, compression belts are known to cause a high level of patient discomfort, which can increase the variability of tumor motion [
      • Mampuya W.A.
      • Matsuo Y.
      • Ueki N.
      • Nakamura M.
      • Mukumoto N.
      • Nakamura A.
      • et al.
      The impact of abdominal compression on outcome in patients treated with stereotactic body radiotherapy for primary lung cancer.
      ,
      • Gelder R.
      • Wong S.
      • Le A.
      • Podreka A.
      • Briggs A.
      • Haddad C.
      • et al.
      Experience with an abdominal compression band for radiotherapy of upper abdominal tumours.
      ,
      • Bissonnette J.P.
      • Franks K.V.
      • Purdie T.G.
      • Moseley D.J.
      • Sonke J.J.
      • Jaffray D.A.
      • et al.
      Quantifying interfraction and intrafraction tumor motion in lung stereotactic body radiotherapy using respiration-correlated cone beam computed tomography.
      ].
      As a solution to reduce respiration-induced tumor motion in patients with pancreatic cancer, polyurethane-based customized abdominal corsets have been used in stereotactic RT with photons (XT) [
      • Heerkens H.D.
      • Hall W.A.
      • Li X.A.
      • Knechtges P.
      • Dalah E.
      • Paulson E.S.
      • et al.
      Recommendations for MRI-based contouring of gross tumor volume and organs at risk for radiation therapy of pancreatic cancer.
      ]. Studies using orthogonal 2D-cine or 4D MRI have indicated the ability of such corsets to reduce respiratory-induced pancreatic motion [
      • Heerkens HD
      • Reerink O
      • Intven MPW
      • Hiensch RR
      • van den Berg CAT
      • Crijns SPM
      • et al.
      Pancreatic tumor motion reduction by use of a custom abdominal corset.
      ,
      • Heerkens H.D.
      • Hall W.A.
      • Li X.A.
      • Knechtges P.
      • Dalah E.
      • Paulson E.S.
      • et al.
      Recommendations for MRI-based contouring of gross tumor volume and organs at risk for radiation therapy of pancreatic cancer.
      ,
      • Dolde K
      • Schneider S
      • Stefanowicz S
      • Alimusaj M
      • Flügel B
      • Saito N
      • et al.
      Comparison of pancreatic respiratory motion management with three abdominal corsets for particle radiation therapy: Case study.
      ,
      • Grimbergen G.
      • Eijkelenkamp H.
      • Heerkens H.D.
      • Raaymakers B.W.
      • Intven M.P.W.
      • Meijer G.J.
      Intrafraction pancreatic tumor motion patterns during ungated magnetic resonance guided radiotherapy with an abdominal corset.
      ]. However, corsets used in XT may not be applicable for PT, since for the latter a high setup reproducibility for immobilization devices placed in the beam path is mandatory and more stringent requirements for material properties (i.e. thickness and homogeneity) have to be met. Therefore, in a previous study-three different types of corsets have been developed and investigated in their applicability for a combination of PT and MRI [
      • Dolde K
      • Schneider S
      • Stefanowicz S
      • Alimusaj M
      • Flügel B
      • Saito N
      • et al.
      Comparison of pancreatic respiratory motion management with three abdominal corsets for particle radiation therapy: Case study.
      ]. In the hereby presented study the custom-developed abdominal corset deemed ideal for this application was investigated for the first time in a clinical setting regarding its ability to reduce intrafraction pancreas motion under free-breathing conditions as well as to assess patient tolerability and acceptance.

      Material and methods

      Patients acquisition

      In this clinical study, approved by the local Ethics Committee, patients diagnosed with an unresectable tumor originating from the pancreas, gallbladder or liver, prior to neoadjuvant chemoradiotherapy, with no present distant metastasis and a good general state of health (ECOG performance status 0 or 1) were included in this study after providing a written informed consent.

      Particle therapy compatible abdominal corset

      The abdominal corset presented in this study was co-developed and fabricated by a manufacturer for orthotic and prosthetic systems (Orthopädie- und Rehatechnik Dresden GmbH, Dresden, Germany). The patient-individual manufacturing was based on an optical 3D-surface scan (Artec Eva®, Artec3D, Luxembourg, Luxembourg) of the patient in treatment position using a vacuum cushion (TapMed Medizintechnik Handels GmbH, Habichtswald, Germany) for modeling of the dorsum from its imprint. In the 3D-surface model the stomach size was reduced according to the tolerability of the patient, to particularly create pressure in the abdomen when applied. The resulting surface model was replicated by milling a wooden blank into a respective model onto which a 5 mm polyethylene (PE) plate (Streifylast®/Trolen, Streifeneder ortho.produktion GmbH, Emmering, Germany) was thermoformed in an industrial oven at 150 °C. The corset (Fig. 2) was finally adjusted and equipped with hook-and-pile Velcro fasteners, placed laterally left of the patient so as to not impede with the radiation beams, typically applied from angles between 100° − 360° [
      • Stefanowicz S.
      • Stützer K.
      • Zschaeck S.
      • Jakobi A.
      • Troost E.G.C.
      Comparison of different treatment planning approaches for intensity-modulated proton therapy with simultaneous integrated boost for pancreatic cancer.
      ]. This choice of design ensured that a potential positional variability of the corset would not have a disturbing effect on the PT dose distribution delivered, considering the homogeneous material and constant thickness the corset is made of [
      • Schneider S.
      Inter- and Intrafraction Motion Management for MR guided Proton Therapy of Pancreatic Carcinoma.
      ]. To prevent the fasteners from being covered by the vacuum mattress during positioning of the patient, the buckles had to be positioned anterior of the patient’s mid-plane without colliding with a potential AP proton beam. In IS direction, the corset ranged from the xiphoid process to the ilium, compressing the stomach and the lower five ribs. For the modeling and manufacturing the patient was scheduled for two appointments of approximately 20 min duration each and a total manufacturing time of at least 3 days between the two appointments. A recess of 10×6 cm2 was incorporated into the corset at the lower right of the abdomen caudal to the umbilicus, which facilitated the use of a pneumatic sensor during acquisition of the treatment planning CT scan which is used for 4D-CT image reconstruction. The PE-based abdominal corset was fabricated exclusively from MR-safe materials. Furthermore, its thin design allowed the receiver coils to be placed close to the patient’s body, thereby preventing a loss in SNR and hence MR image quality deterioration.

      MRI acquisition and pancreas motion analysis

      For the quantification of respiration-induced pancreatic motion and its reduction through the use of the abdominal corset, patients were scanned in supine position on a flat table-top overlay (Medibord Ltd., Nottingham, United Kingdom) with a 3.0 T MRI scanner (Philips Ingenuity TF PET/MR scanner, Philips Healthcare, Eindhoven, The Netherlands) using posterior and anterior 32-channel SENSE Torso/Cardiac receiver coils. These coils were positioned around the patient using a dedicated coil holder system (Medibord Ltd.) to avoid additional pressure from the coils on the body that could affect the breathing behavior (Fig. 1). Unlike the conventionally used arms-up positioning during RT, in this study the arms were positioned next to the body due to the limited space in the 60 cm diameter bore of the MRI scanner, which was further on reduced in size by the table-top overlay and coil holder system. MR imaging was first performed while the patient was wearing the corset. Immediately thereafter, the corset was removed and the patient was re-scanned to acquire a reference motion dataset. No breathing instructions were given during any of the scans.
      Figure thumbnail gr1
      Fig. 1MRI scanner (3 T Philips Ingenuity TF PET/MR scanner) modified for mimicking RT conditions by addition of a flat table-top overlay and a coil holder system for both the anterior and posterior receiver coils.
      Figure thumbnail gr2
      Fig. 2Front view of a patient wearing the abdominal corset. A recess is incorporated into the corset (a) for positioning a respiratory sensor (b) used for 4D CT image reconstruction.
      For the assessment of the mean pancreas motion as well as its intrafraction variability, orthogonal single-slice 2D-cine MRI as well as phase resolved 4D MRI were evaluated, due to the partially complementary information that the 2D-cine and 4D MR images provide. 2D-cine MRI was acquired sequentially in coronal and sagittal orientation at a frame rate of 2 Hz (field of view = 375×375 mm2, resolution = 1.0×1.0 mm2, slice thickness = 7 mm, bandwidth = 1929 Hz/pixel, TE = 1.1 ms, TR = 2.2 ms, flip angle = 27°, acquisition time = 1 min). After image acquisition, the pancreas in all dynamics was manually segmented for all patients by the same observer and its center-of-mass motion was characterized in inferior-superior (IS), anterior-posterior (AP) and left–right (LR) directions by the mean motion amplitude (M95%) and its intrafraction variability. IS motion information was thereby extracted from acquisition in both sagittal and coronal planes and reported as the mean of both. The M95% represents 95 % of the bandwidth of all data points, excluding the 2.5 percentile most extreme data points on both inspiration and expiration to reduce the effect of outliers. The intrafraction variability was thereby defined as the standard deviation of all peak-to-peak motion amplitudes of the respective respiratory cycles.
      Furthermore, 4D MRI datasets were reconstructed to obtain an averaged 3D motion assessment of the full pancreas volume by retrospectively binning a multi-slice 2D-cine MRI acquisition (field of view = 375×375×120 mm3, resolution = 1.0×1.0 mm2, slice thickness = 5 mm, bandwidth = 1929 Hz/pixel, TE = 1.1 ms, TR = 2.2 ms, flip angle = 27°, acquisition time = 9 min) to 10 respiratory phases with differentiation between mid-inhale and mid-exhale phases. The resorting algorithm was based on the method described by Von Siebenthal et al. [
      • Siebenthal M.
      • Székely G.
      • Gamper U.
      • Boesiger P.
      • Lomax A.
      • Cattin P.H.
      4D MR imaging of respiratory organ motion and its variability.
      ]. However, instead of acquiring a navigator slice in between every-two image slices, the reconstruction in this study was performed with an image-based navigation method. There, the continuously acquired parallel coronal 2D images of a 3D volume with 50 dynamics were resorted based on the position of the dome of the liver in each MR image, as a proxy for the respiratory state. On each 4D MRI data set, the pancreas was manually delineated on the end-inhale and end-exhale images using the open-source software MITK (German Cancer Research Center, Heidelberg, Germany). The center-of-mass of the respective delineations was determined to extract the mean pancreatic motion M100% in IS, AP and LR directions. Since contrary to the 2D-cine data only one respiratory cycle was reconstructed, no segmented data was excluded as outliers. The statistical significance of measured motion reduction by use of the abdominal corset was analyzed based on a paired sample t-test.

      Patient comfort under application of an abdominal corset

      In order to analyze patient comfort when applying of the corset, the patients were asked to complete a questionnaire assessing the level of discomfort in various regions and degrees after its application, both during corset manufacturing and during the acquisition of both the treatment planning CT and the MRI scans. Patients wore the corset for approximately 10 min during manufacturing, and 30 min during both CT and MRI scanning, respectively.

      Results

      Patient characteristics and data inclusion

      Sixteen patients (5 male and 11 female; average ± SD age 70.1 ± 9.0 years) provided written informed consent for the manufacturing of a patient-individualized abdominal corset and for acquiring MRI scans with pulse sequences optimized for motion characterization. Out of these 16 patients initially included in the study, in one patient the MRI scan had to be aborted due to claustrophobia, while three other patients did not fit into the bore of MR scanner with the anterior receiver coil being positioned onto the coil holder. These patients could therefore only be scanned without coil holder, with the anterior receiver coil positioned in direct contact with their torso. Hence, since no reference scan could be acquired, their MRI data was not included in the motion analysis. Thus, for the motion analysis, data from 12 patients were available (3 male and 9 female; average ± SD age 70.6 ± 8.8 years, see Table 1). Out of the 12 patient datasets, for two patients the reconstruction of the 4D MRI dataset (initially based on radial k-space sampling) did not result in images of sufficient quality for analysis. Consequentially, 4D acquisition was changed to a multi-slice based 4D MR pulse sequence, where during the adaptation of the MR protocol for one patient erroneously no orthogonal 2D-cine MRI dataset was acquired. Hence, 11 patients were evaluated using 2D-cine MRI, and 10 patients were evaluated using 4D MRI.
      Table 1Detailed characteristics of the 12 patients included in the pancreas motion analysis.
      Patient #SexAgeWeightTumor locationTumor type
      1Male5167.7Bile ductCholangiocarcinoma, Recurrence
      2Male6769.8Pancreatic tailAdenocarcinoma, Malignant neoplasm
      3Female7963.1Pancreatic headAdenocarcinoma, Malignant neoplasm
      4Female6560.4Pancreas multiple areas overlappingAdenocarcinoma, Malignant neoplasm
      5Female8461.8Pancreatic headAdenocarcinoma, Malignant neoplasm
      6Female7745.0Pancreatic headAdenocarcinoma, Malignant neoplasm
      7Female8068.8LiverHepatocellular carcinoma, Malignant neoplasm
      8Male7671.4Pancreatic headAdenocarcinoma, Malignant neoplasm
      9Female6564.0Pancreatic bodyAdenocarcinoma, Recurrence
      10Female7257.3Pancreas multiple areas overlappingAdenocarcinoma, Malignant neoplasm
      11Female6660.3LiverMantle cell lymphoma, Recurrence
      12Female6566.1Pancreatic headAdenocarcinoma, Malignant neoplasm

      Pancreatic motion evaluation

      Both the orthogonal 2D-cine and the 4D MRI data showed that the use of the corset led to a reduction in pancreatic center-of-mass motion, mainly in the IS direction (Fig. 3). Without corset, the analysis of the 2D-cine MRI revealed a population-averaged M95% in IS direction of 6.9 mm ± 3.1 mm, which was significantly reduced by 47 % (p < 0.05) to 3.7 mm ± 1.5 mm when using the corset. The M95% in AP and LR direction without corset was substantially smaller than in IS direction with 1.5 mm ± 0.5 mm in AP and 1.7 mm ± 1.1 mm in LR direction. Through the application of the corset, the M95% in these directions was reduced to 1.0 mm ± 0.6 mm in AP and to 1.3 mm ± 0.9 mm in LR direction, albeit not statistically significant. Furthermore, the intrafraction motion variability significantly decreased in IS direction by 35.8 % (p < 0.01) from 1.3 mm ± 0.4 mm to 0.8 mm ± 0.3 mm when the corset was applied.
      Figure thumbnail gr3
      Fig. 3Box-plots showing the measured motion amplitude of the pancreas as center-of-mass in inferior–superior (IS), anterior–posterior (AP) and left–right (LR) direction with and without the corset applied, extracted from [left] 2D-cine MRI data and [right] retrospectively resorted 4D MRI data. Pancreas motion was characterized as M95% and M100% for 2D-cine MRI and 4D MRI, respectively. The whiskers include all data points, with no outliers disregarded.
      On 4D MRI, the results found on 2D-cine MRI were echoed in that the corset reduced pancreatic center-of-mass motion significantly in IS direction by 51 % (from on average 9.3 mm ± 4.5 mm to 4.6 mm ± 2.1 mm, p < 0.05) while it did not significantly reduce motion in the AP and LR directions (AP: 2.1 mm ± 2.4 mm to 1.6 mm ± 0.9 mm; LR: 3.1 mm ± 2.2 mm to 2.7 mm ± 1.5 mm).

      Patient comfort

      All 16 patients completed the questionnaire (supplementary Table S1). The corset was generally well tolerated with on average no complaints nor pain in any category (Fig. 4). Only two patients who already suffered from back pain before fabrication of the corset mentioned light or strong back pain. All 16 patients wore their individualized corset during all fractions of their radiation treatment.
      Figure thumbnail gr4
      Fig. 4Stacked bar graph showing the result of the patient survey, analyzing their acceptance towards application of the customized abdominal corset.

      Discussion

      Particularly in proton therapy accurate tumor positioning is essential, due to its sensitivity to density variations along the beam path. In patients with pancreatic tumors, accurate tumor positioning is challenged by respiratory induced tumor motion where motion amplitudes of 15 – 25 mm have been observed [
      • Feng M.
      • Balter J.M.
      • Normolle D.
      • Adusumilli S.
      • Cao Y.
      • Chenevert T.L.
      • et al.
      Characterization of pancreatic tumor motion using cine MRI: surrogates for tumor position should be used with caution.
      ,
      • Heerkens H.D.
      • van Vulpen M.
      • van den Berg C.A.T.
      • Tijssen R.H.N.
      • Crijns S.P.M.
      • Molenaar I.Q.
      • et al.
      MRI-based tumor motion characterization and gating schemes for radiation therapy of pancreatic cancer.
      ]. In order to better exploit the potential dosimetric benefit of PT, an MR-guided PT workflow for patients with pancreatic cancer was established which includes a motion mitigation method. For motion mitigation a patient-individualized MRI- and PT-compatible abdominal corset was first applied in patients after evaluation of several candidate corsets in a preceding feasibility study [
      • Dolde K
      • Schneider S
      • Stefanowicz S
      • Alimusaj M
      • Flügel B
      • Saito N
      • et al.
      Comparison of pancreatic respiratory motion management with three abdominal corsets for particle radiation therapy: Case study.
      ]. This corset was then selected and adjusted to comply with our in-house clinical workflow. Its performance to reduce respiratory-induced pancreas motion and patient tolerability was investigated for the first time in this clinical pilot study.
      The treatment planning of moving tumors in our in-house clinical routine is based on 4D CT data, which is retrospectively reconstructed by utilizing the respiratory signal acquired with a pressure sensor (Anzai Medical Co., Ltd., Tokyo, Japan) attached on a fixation belt placed around the patient’s abdomen. Positioning of the sensor superior to the corset on the patient’s upper right chest or on the stomach covered by the corset led to a poor performance of the sensor. The recess of 10×6 cm2 incorporated into the caudal part of the corset allowed for an extraction of the respiratory signal while still applying pressure to the patient’s stomach.
      The effect of the abdominal corset in reducing breathing-induced pancreas motion found in this study is in line with studies found in literature analyzing the pancreas motion reduction by a different type of corset. In [
      • Heerkens H.D.
      • Hall W.A.
      • Li X.A.
      • Knechtges P.
      • Dalah E.
      • Paulson E.S.
      • et al.
      Recommendations for MRI-based contouring of gross tumor volume and organs at risk for radiation therapy of pancreatic cancer.
      ] an RT-compatible polyurethane corset was investigated in a patient study (n = 10), where pancreas motion was reduced through the corset in IS direction from an average 11.3 mm to 7.2 mm but without significant effect in AP and LR direction. Considering the moderately sized patient cohort and the large inter-patient variability in pancreas motion found, the results correspond well with those found in the current study. A material analysis of, amongst others, a polyurethane-based corset [
      • Dolde K
      • Schneider S
      • Stefanowicz S
      • Alimusaj M
      • Flügel B
      • Saito N
      • et al.
      Comparison of pancreatic respiratory motion management with three abdominal corsets for particle radiation therapy: Case study.
      ] found this type of corset to be potentially incompatible with PT due to material inhomogeneities and large variabilities in thickness. In [
      • Grimbergen G.
      • Eijkelenkamp H.
      • Heerkens H.D.
      • Raaymakers B.W.
      • Intven M.P.W.
      • Meijer G.J.
      Intrafraction pancreatic tumor motion patterns during ungated magnetic resonance guided radiotherapy with an abdominal corset.
      ] a polyurethane-based corset was again investigated in a patient study (n = 13) by analyzing pancreas motion only after compression, without reference measurements acquired without corset. Pancreas motion after abdominal compression was found to be small, with the most prominent motion being found in the IS direction at 4.2 mm and smaller amplitudes of 2.3 mm and 1.4 mm in AP and LR directions, respectively, corresponding to the reduced motion found in this study.
      In this study, for the first time, a systematic examination of patient tolerability and acceptance for wearing an abdominal corset during fabrication and treatment simulation was performed. While 14 out of 16 patients reported no pain related to the application of the corset, two patients reported light or strong back pain. Since both patients suffered from most likely tumor-related back pain already before application of the corset, we considered these as outliers not related to the corset.
      However, several limitations of the current study need to be considered and discussed. First, due to the small space remaining in the bore of the MR scanner due to the placement of the flat table-top overlay and receiver coil holder system, patients could not be positioned with their arms above their head as is commonly performed in abdominal RT. This difference in arm positioning could have an altering effect on breath-induced motion amplitude. In addition, the limited space inside the MR scanner also led to a biased patient acquisition, including only patients having a small abdomen and thorax with a maximum AP extension of 27 cm.
      The patients included in this study were scanned with both orthogonal 2D-cine MRI as well as retrospectively resorted 4D MRI, due to the partially complementary information these methods provided. The pancreas motion extracted from both methods revealed a similar reduction in pancreatic motion, mainly in the IS direction, by 47 % and 51 %, respectively. Yet, the absolute motion amplitudes differed marginally, where 4D MRI showed larger motion amplitudes in both uncompressed as well as compressed breathing. This divergence is likely affected by several factors. First, single-slice 2D-cine MRI only captures the motion of a selected slice, not accurately covering potential 3D organ deformation. Furthermore, the apparent motion in 2D-cine images can be affected by through-plane motion, which can distort the real motion. On the other hand, contrary to 4D-MRI, 2D-cine MRI does not suffer from re-binning errors due to its real-time imaging nature. 4D MRI includes the full volume of the pancreas and is hence not affected by through-plane motion. On the other hand, the reconstruction algorithm used for 4D MRI can potentially outweigh deep-inspiration states, which would potentially be discarded as outliers in the M95% motion analysis performed on 2D-cine MRI. No data was discarded in the 4D MRI reconstruction and for every slice the maximum state of inspiration and expiration was determined individually based on the cranio-caudal position of the dome of the liver. This effect is partially compensated by the larger volume over which the pancreas motion was determined in 4D MRI but could influence the results towards larger motion amplitudes. Additionally, the poorer image quality of the reconstructed 4D MR images could have degraded the segmentation accuracy of the pancreas. Since the breathing induced pancreas motion was found to be systematically larger in the 4D MRI data than in the 2D-cine MRI data throughout the whole patient cohort, segmentation uncertainty cannot explain the observed difference in absolute pancreas motion between 2D-cine and 4D MRI.
      On 4D MRI, the mean pancreas motion in the IS direction without compression was found to be 9.3 mm ± 4.5 mm, showing the need for motion mitigation for this indication. By use of the abdominal corset, this motion was reduced to 4.6 mm ± 2.1 mm, where a motion amplitude of <5 mm is commonly suggested as threshold at which no further motion mitigation is necessary for some radiotherapy applications [
      • Keall P.J.
      • Mageras G.S.
      • Balter J.M.
      • Emery R.S.
      • Forster K.M.
      • Jiang S.B.
      • et al.
      The management of respiratory motion in radiation oncology report of the AAPM Task Group 76.
      ]. However, a large interpatient variance was found with 45 % of the patients already showing pancreatic motion of <5 mm without abdominal compression. Therefore, based on the results of the current study, it is recommended that an initial analysis of respiratory-induced motion would be performed prior to applying abdominal compression methods.

      Conclusions

      A custom developed MRI- and PT-compatible abdominal corset, presented in a preceding feasibility study, was successfully integrated into our in-house clinical workflow for PT of tumors of the upper abdomen and used for the first time in a clinical study. The corset substantially reduces breathing-induced pancreas motion and can be safely applied with no additional discomfort for the patients.

      Declaration of Competing Interest

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Acknowledgements

      The authors thank all patients agreeing to participate in the presented study. Furthermore, we thank the colleagues from the Department of Nuclear Medicine of the University Hospital Carl Gustav Carus Dresden for their support when imaging on the PET/MRI scanner.

      Funding

      The production of the abdominal corsets was in part funded by each, the German Cancer Research Center (DKFZ) and Helmholtz-Zentrum Dresden-Rossendorf (HZDR).

      Clinical trial information

      This clinical study was approved by the local Ethics Committee of the Faculty of Medicine and University Hospital Carl Gustav Carus of the Technische Universität Dresden (EK-398092016).

      Appendix A. Supplementary data

      The following are the Supplementary data to this article:

      References

        • Hiroshima Y.
        • Fukumitsu N.
        • Saito T.
        • Numajiri H.
        • Murofushi K.N.
        • Ohnishi K.
        • et al.
        Concurrent chemoradiotherapy using proton beams for unresectable locally advanced pancreatic cancer.
        Radiother Oncol. 2019; 136: 37-43https://doi.org/10.1016/j.radonc.2019.03.012
        • Hong T.S.
        • Ryan D.
        • Borger D.
        • Blaszkowsky L.S.
        • Yeap B.Y.
        • Ancukiewicz M.
        • et al.
        A phase 1/2 and biomarker study of preoperative short course chemoradiation with proton beam therapy and capecitabine followed by early surgery for resectable pancreatic ductal adenocarcinoma.
        Int J Radiat Oncol Biol Phys. 2014; 89: 830-838https://doi.org/10.1016/j.ijrobp.2014.03.034
        • Jethwa K.R.
        • Tryggestad E.J.
        • Whitaker T.J.
        • Giffey B.T.
        • Kazemba B.D.
        • Neben-Wittch M.A.
        • et al.
        Initial experience with intensity modulated proton therapy for intact, clinically localized pancreas cancer: Clinical implementation, dosimetric analysis, acute treatment-related adverse events, and patient-reported outcomes.
        Adv Radiat Oncol. 2018; 3: 314-321https://doi.org/10.1016/j.adro.2018.04.003
        • Kim T.H.
        • Lee W.J.
        • Woo S.M.
        • Oh E.S.
        • Youn S.H.
        • Jang H.Y.
        • et al.
        Efficacy and feasibility of proton beam radiotherapy using the simultaneous integrated boost technique for locally advanced pancreatic cancer.
        Sci Rep. 2020; 10: 21712https://doi.org/10.1038/s41598-020-78875-1
        • Rutenberg MS
        • Nichols RC
        Proton beam radiotherapy for pancreas cancer.
        J Gastrointest Oncol. 2020; 11: 166-175https://doi.org/10.21037/jgo.2019.03.02
        • Reese A.S.
        • Lu W.
        • Regine W.F.
        Utilization of intensity-modulated radiation therapy and image-guided radiation therapy in pancreatic cancer: is it beneficial?.
        Semin Radiat Oncol. 2013; 24: 132-139https://doi.org/10.1016/j.semradonc.2013.11.003
        • Bert C.
        • Grözinger S.O.
        • Rietzel E.
        Quantification of interplay effects of scanned particle beams and moving targets.
        Phys Med Biol. 2008; 53: 2253-2265https://doi.org/10.1088/0031-9155/53/9/003
        • Phillips M.H.
        • Pedroni E.
        • Blattmann H.
        • Boehringer T.
        • Coray A.
        • Scheib S.
        Effects of respiratory motion on dose uniformity with a charged particle scanning method.
        Phys Med Biol. 1992; 37: 223-233https://doi.org/10.1088/0031-9155/37/1/016
        • Graeff C.
        Motion mitigation in scanned ion beam therapy through 4D-optimization.
        Phys Med. 2014; 30: 570-577https://doi.org/10.1016/j.ejmp.2014.03.011
        • Bert C.
        • Durante M.
        Motion in radiotherapy: particle therapy.
        Phys Med Biol. 2011; 56 (R144): R113https://doi.org/10.1088/0031-9155/56/16/r01
        • Dolde K
        • Dávid C
        • Echner G
        • Floxa R
        • Hentschke C
        • Maier F
        • et al.
        4DMRI-based analysis of inter- and intrafractional pancreas motion and deformation with different immobilization devices.
        Biomed Phys Eng Express. 2019; 5025012https://doi.org/10.1088/2057-1976/aaf9ae
        • Knopf A.C.
        • Hong T.S.
        • Lomax A.
        Scanned proton radiotherapy for mobile targets the effectiveness of re-scanning in the context of different treatment planning approaches and for different motion characteristics.
        Phys Med Biol. 2011; 56: 7257-7271https://doi.org/10.1088/0031-9155/56/22/016
        • Bert C.
        • Saito N.
        • Schmidt A.
        • Chaudhri N.
        • Scardt D.
        • Rietzel E.
        Target motion tracking with a scanned particle beam.
        Med Phys. 2007; 34: 4768-4771https://doi.org/10.1118/1.2815934
        • Mori S.
        • Yanagi T.
        • Hara R.
        • Sharp G.C.
        • Asakura H.
        • Kumagai,
        • et al.
        Comparison of respiratory-gated and respiratory-ungated planning in scattered carbon ion beam treatment of the pancreas using four-dimensional computed tomography.
        Int J Radiat Oncol Biol Phys. 2010; 76: 303-312https://doi.org/10.1016/j.ijrobp.2009.05.026
        • Yamada T.
        • Takao S.
        • Koyano H.
        • Nihongi H.
        • Fujii Y.
        • Hirayama S.
        • et al.
        Validation of dose distribution for liver tumors treated with real-time-image gated spot-scanning proton therapy by log data based dose reconstruction.
        J Radiat Res. 2021; 62: 626-633https://doi.org/10.1093/jrr/rrab024
        • Kraus K.M.
        • Heath E.
        • Oelfke U.
        Dosimetric consequences of tumour motion due to respiration for a scanned proton beam.
        Phys Med Biol. 2011; 56: 6563-6581https://doi.org/10.1088/0031-9155/56/20/003
        • Zhang Y.
        • Huth I.
        • Weber D.C.
        • Lomax A.J.
        A statistical comparison of motion mitigation performances and robustness of various pencil beam scanned proton systems for liver tumour treatments.
        Radiother Oncol. 2018; 128: 182-188https://doi.org/10.1016/j.radonc.2018.01.019
        • Yamazaki H.
        • Nishiyama K.
        • Tanaka E.
        • Koiwai K.
        • Shikama N.
        • Yoshinori I.
        • et al.
        Dummy run for a phase II multi-institute trial of chemoradiotherapy for unresectable pancreatic cancer: inter-observer variance in contour delineation.
        Anticancer Res. 2007; 27 (https://ar.iiarjournals.org/content/27/4C/2965): 2965-2971
        • Fontana G
        • Riboldi M
        • Gianoli C
        • Chirvase CI
        • Villa G
        • Paganelli Chiara
        • et al.
        MRI quantification of pancreas motion as a function of patient setup for particle therapy —a preliminary study.
        J Appl Clin Med Phys. 2016; 17: 60-75https://doi.org/10.1120/jacmp.v17i5.6236
        • Heerkens HD
        • Reerink O
        • Intven MPW
        • Hiensch RR
        • van den Berg CAT
        • Crijns SPM
        • et al.
        Pancreatic tumor motion reduction by use of a custom abdominal corset.
        Phys Imaging Radiat Oncol. 2017; 2: 7-10https://doi.org/10.1016/j.phro.2017.02.003
        • Dolde K.
        • Naumann P.
        • Dávid C.
        • Gnirs R.
        • Kachelrieß M.
        • Lomax A.J.
        • et al.
        4D dose calculation for pencil beam scanning proton therapy of pancreatic cancer using repeated 4DMRI datasets.
        Phys Med Biol. 2018; 63165005https://doi.org/10.1088/1361-6560/aad43f
        • Wroe A.J.
        • Bush A.D.
        • Slater J.D.
        Immobilization Considerations for Proton Radiation Therapy.
        Technol Cancer Res Treat. 2014; 13: 217-226https://doi.org/10.7785/tcrt.2012.500376
        • Newhauser WD
        • Zhang R
        The physics of proton therapy.
        Phys Med Biol. 2015; 60: R155-R209https://doi.org/10.1088/0031-9155/60/8/R155
        • Mampuya W.A.
        • Matsuo Y.
        • Ueki N.
        • Nakamura M.
        • Mukumoto N.
        • Nakamura A.
        • et al.
        The impact of abdominal compression on outcome in patients treated with stereotactic body radiotherapy for primary lung cancer.
        J Radiat Res. 2014; 55: 934-939https://doi.org/10.1093/jrr/rru028
        • Gelder R.
        • Wong S.
        • Le A.
        • Podreka A.
        • Briggs A.
        • Haddad C.
        • et al.
        Experience with an abdominal compression band for radiotherapy of upper abdominal tumours.
        J Med Radiat Sci. 2018; 65: 48-54https://doi.org/10.1002/jmrs.254
        • Bissonnette J.P.
        • Franks K.V.
        • Purdie T.G.
        • Moseley D.J.
        • Sonke J.J.
        • Jaffray D.A.
        • et al.
        Quantifying interfraction and intrafraction tumor motion in lung stereotactic body radiotherapy using respiration-correlated cone beam computed tomography.
        Int J Radiat Oncol Biol Phys. 2009; 75: 688-695https://doi.org/10.1016/j.ijrobp.2008.11.066
        • Heerkens H.D.
        • Hall W.A.
        • Li X.A.
        • Knechtges P.
        • Dalah E.
        • Paulson E.S.
        • et al.
        Recommendations for MRI-based contouring of gross tumor volume and organs at risk for radiation therapy of pancreatic cancer.
        Pract Radiat Oncol. 2017; 7: 126-136https://doi.org/10.1016/j.prro.2016.10.006
        • Dolde K
        • Schneider S
        • Stefanowicz S
        • Alimusaj M
        • Flügel B
        • Saito N
        • et al.
        Comparison of pancreatic respiratory motion management with three abdominal corsets for particle radiation therapy: Case study.
        J Appl Clin Med Phys. 2019; 20: 111-119https://doi.org/10.1002/acm2.12613
        • Grimbergen G.
        • Eijkelenkamp H.
        • Heerkens H.D.
        • Raaymakers B.W.
        • Intven M.P.W.
        • Meijer G.J.
        Intrafraction pancreatic tumor motion patterns during ungated magnetic resonance guided radiotherapy with an abdominal corset.
        Phys Imaging Radiat Oncol. 2021; 21: 1-5https://doi.org/10.1016/j.phro.2021.12.001
        • Siebenthal M.
        • Székely G.
        • Gamper U.
        • Boesiger P.
        • Lomax A.
        • Cattin P.H.
        4D MR imaging of respiratory organ motion and its variability.
        Phys Med Biol. 2007; 52: 1547-1564https://doi.org/10.1088/0031-9155/52/6/001
        • Stefanowicz S.
        • Stützer K.
        • Zschaeck S.
        • Jakobi A.
        • Troost E.G.C.
        Comparison of different treatment planning approaches for intensity-modulated proton therapy with simultaneous integrated boost for pancreatic cancer.
        Radiat Oncol. 2018; 13: 228https://doi.org/10.1186/s13014-018-1165-0
        • Schneider S.
        Inter- and Intrafraction Motion Management for MR guided Proton Therapy of Pancreatic Carcinoma.
        Technische Universität Dresden, 2020 (PhD Dissertation,)
        • Feng M.
        • Balter J.M.
        • Normolle D.
        • Adusumilli S.
        • Cao Y.
        • Chenevert T.L.
        • et al.
        Characterization of pancreatic tumor motion using cine MRI: surrogates for tumor position should be used with caution.
        Int J Radiat Oncol Biol Phys. 2009; 74: 884-891https://doi.org/10.1016/j.ijrobp.2009.02.003
        • Heerkens H.D.
        • van Vulpen M.
        • van den Berg C.A.T.
        • Tijssen R.H.N.
        • Crijns S.P.M.
        • Molenaar I.Q.
        • et al.
        MRI-based tumor motion characterization and gating schemes for radiation therapy of pancreatic cancer.
        Radiother Oncol. 2014; 111: 252-257https://doi.org/10.1016/j.radonc.2014.03.002
        • Keall P.J.
        • Mageras G.S.
        • Balter J.M.
        • Emery R.S.
        • Forster K.M.
        • Jiang S.B.
        • et al.
        The management of respiratory motion in radiation oncology report of the AAPM Task Group 76.
        Med Phys. 2006; 33: 3874-3900https://doi.org/10.1118/1.2349696