A Systematic Review of the Clinical Implementation of Pelvic Magnetic Resonance Imaging–Only Planning for External Beam Radiation Therapy

Open AccessPublished:July 01, 2019DOI:https://doi.org/10.1016/j.ijrobp.2019.06.2530
      The use of magnetic resonance (MR) imaging scans alone for radiation therapy treatment planning (MR-only planning) has been highlighted as one method of improving patient outcomes. Recent technologic advances have meant that introducing MR-only planning to the clinic is becoming a reality, with several specialist radiation therapy clinics using this technique for treatment. As such, substantial efforts are being made to introduce this technique into wide-spread clinical implementation. A systematic review of publications investigating the clinical implementation of pelvic MR-only radiation therapy treatment planning was undertaken following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The Medline, Embase, Scopus, Science Direct, Cumulative Index to Nursing and Allied Health Literature, and Web of Science databases were searched (timespan: all years to January 2, 2019). Twenty-six articles met the inclusion criteria. The studies were grouped into the following categories: (1) MR acquisition and synthetic computed tomography generation verification, (2) MR distortion quantification and phantom development, (3) clinical validation of patient treatment positioning in an MR-only workflow, and (4) MR-only commissioning processes. Key conclusions from this review are (1) MR-only planning has been implemented clinically for prostate cancer treatments; (2) a substantial amount of work remains to translate MR-only planning into widespread clinical implementation for all pelvic sites; (3) MR scanner distortions are no longer a barrier to MR-only planning, but they must be managed appropriately; (4) MR-only–based patient positioning verification shows promise, but limited evidence is reported in the literature and further investigation is required; and (5) a number of MR-only commissioning processes have been reported, which can aid centers as they undertake local commissioning; however, this needs to be formalized in guidance from national bodies.

      Introduction

      One of greatest challenges remaining in radiation therapy is improving the accuracy of treatment volume delineations.
      • Njeh C.F.
      Tumor delineation: The weakest link in the search for accuracy in radiotherapy.
      Further reducing the uncertainty in delineation could lead to improved patient outcomes by reducing treatment volumes, allowing a reduction in treatment-related toxicities,
      • Sander L.
      • Langkilde N.C.
      • Holmberg M.
      • Carl J.
      MRI target delineation may reduce long-term toxicity after prostate radiotherapy.
      • Wortel R.C.
      • Heemsbergen W.D.
      • Smeenk R.J.
      • et al.
      Local Protocol Variations for Image Guided Radiation Therapy in the Multicenter Dutch Hypofractionation (HYPRO) Trial: Impact of rectal balloon and MRI delineation on anorectal dose and gastrointestinal toxicity levels.
      or reducing the risk of geographic misses, thereby improving local control and potentially overall survival rates. The use of magnetic resonance imaging (MRI) scans alone for radiation therapy treatment planning (magnetic resonance [MR]-only planning) has been highlighted as one method of potentially improving target volume delineation accuracy
      • Johnstone E.
      • Wyatt J.J.
      • Henry A.M.
      • et al.
      Systematic review of synthetic computed tomography generation methodologies for use in magnetic resonance imaging-only radiation therapy.
      • Owrangi A.M.
      • Greer P.B.
      • Glide-Hurst C.K.
      MRI-only treatment planning: Benefits and challenges.
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      ; this is due to the improved soft tissue contrast of MRI compared with computed tomography (CT) and the potential to use the other benefits of MRI, such as functional imaging.
      • Owrangi A.M.
      • Greer P.B.
      • Glide-Hurst C.K.
      MRI-only treatment planning: Benefits and challenges.
      Recent technological advances have meant that introducing MR-only planning to the clinic is becoming a reality.
      • Owrangi A.M.
      • Greer P.B.
      • Glide-Hurst C.K.
      MRI-only treatment planning: Benefits and challenges.
      • Karlsson M.
      • Karlsson M.G.
      • Nyholm T.
      • Amies C.
      • Zackrisson B.
      Dedicated magnetic resonance imaging in the radiotherapy clinic.
      Hardware and software developments have improved the geometric distortion inherent within MR images to levels that are acceptable for radiation therapy treatment planning,
      • Weygand J.
      • Fuller C.D.
      • Ibbott G.S.
      • et al.
      Spatial precision in magnetic resonance imaging-guided radiation therapy: The role of geometric distortion.
      and substantial progress has been made in acquiring electron density information from MRI data alone through synthetic CT generation methods.
      • Johnstone E.
      • Wyatt J.J.
      • Henry A.M.
      • et al.
      Systematic review of synthetic computed tomography generation methodologies for use in magnetic resonance imaging-only radiation therapy.
      • Edmund J.M.
      • Nyholm T.
      A review of substitute CT generation for MRI-only radiation therapy.
      The field of synthetic CT generation has been reviewed,
      • Johnstone E.
      • Wyatt J.J.
      • Henry A.M.
      • et al.
      Systematic review of synthetic computed tomography generation methodologies for use in magnetic resonance imaging-only radiation therapy.
      • Edmund J.M.
      • Nyholm T.
      A review of substitute CT generation for MRI-only radiation therapy.
      and commercial solutions are available, including several prostate solutions and, recently released, a solution for the whole pelvis.
      • Siversson C.
      • Nordström F.
      • Nilsson T.
      • et al.
      Technical note: MRI only prostate radiotherapy planning using the statistical decomposition algorithm.
      • Köhler M.
      • Vaara T.
      • Van Grootel M.
      • et al.
      MR-Only Simulation for Radiotherapy Planning [white paper].
      Consequently, MR-only treatments are being conducted by specialist radiation therapy clinics and over time are likely to move to more widespread clinical implementation.
      This systematic review assesses the literature surrounding the clinical implementation of pelvic MR-only radiation therapy treatment planning with the aim of detailing and discussing the breadth of work that has been undertaken. This review considers only work that has been published in relation to MR-only planning for pelvic external beam radiation therapy.

       Methods and materials

      A systematic review of publications investigating the clinical implementation of pelvic MR-only external beam radiation therapy treatment planning was performed using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.
      • Moher D.
      • Liberati A.
      • Tetzlaff J.
      • Altman D.G.
      PRISMA Group
      Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement.
      The Medline, Embase, Scopus, Science Direct, Cumulative Index to Nursing and Allied Health Literature, and Web of Science databases were searched with a time span of all years to January 2, 2019 (for Medline and Embase, this corresponded to “Week 3 December 2018” and “Week 52 2018,” respectively) using the search protocols in Appendix E1 (available online at https://doi.org/10.1016/j.ijrobp.2019.06.2530).
      Articles were included that referred to “MR-only” or “synthetic-CT” and “radiotherapy” or synonyms of these terms in their title or abstract. These deliberately broad search criteria were used to minimize the risk of relevant studies not being identified. The search results for each database were combined, and duplicates were removed. The remaining results were screened from their titles for eligibility. Primary screening included only search results that were related to the use of MRI in radiation therapy for cancer treatment. Secondary screening included only articles related to the clinical implementation of MR-only external beam radiation therapy treatment planning for pelvic cancer sites. Articles focusing on the general use of MRI in radiation therapy, MRI in brachytherapy, synthetic CT model generation techniques, target volume delineation using MRI, MR image registration, positron emission tomography/MRI in radiation therapy, MRI safety, MR-only contouring, and MR-only fiducial marker identification were excluded. Articles regarding synthetic CT model generation techniques were specifically excluded because they were appraised recently in the literature within 2 review articles
      • Johnstone E.
      • Wyatt J.J.
      • Henry A.M.
      • et al.
      Systematic review of synthetic computed tomography generation methodologies for use in magnetic resonance imaging-only radiation therapy.
      • Edmund J.M.
      • Nyholm T.
      A review of substitute CT generation for MRI-only radiation therapy.
      and because this review is focused on the clinical implementation, rather than the technique development, aspect of MR-only planning. Conference proceedings were excluded because of their large number and variable information provision, which made their inclusion unbeneficial. A backward citation search of the remaining eligible studies was undertaken. The included studies were categorized according to their focus. For each category, key findings from each study were included in a data table.

      Results

      The database search results can be seen in Figure 1. The combined database search resulted in 2024 records, with 1066 records remaining after duplicate removal. After primary screening, 535 records remained. After secondary screening, 71 studies remained. After further review, 49 studies were removed (44 conference abstracts and 5 studies that did not meet the eligibility criteria). Twenty-two studies remained, to which the citation search added 4 studies; thus, 26 studies were included in this systematic review. The categories and number of excluded articles can be seen in Table 1.
      Figure thumbnail gr1
      Fig. 1Flowchart of the systematic review process, including the number of studies included in this review.
      Table 1Categories and number of articles excluded from this review after primary, secondary, and tertiary screening
      Reasons for exclusionNo. of articles
      Primary screening
       Not related to cancer treatment65
       Not related to radiation therapy cancer treatment301
       Not related to the use of MRI in radiation therapy cancer treatment165
       Total531
      Secondary screening
       General use of MRI115
       Other site synthetic CT generation technique98
       Brain synthetic CT generation technique65
       Brachytherapy, Gamma knife48
       Prostate synthetic CT generation technique43
       MR delineation33
       MR image registration12
       Proton or ion synthetic CT generation technique12
       Other10
       MR-only contouring7
       MR-only fiducial marker identification7
       PET-MRI7
       UTE synthetic CT generation techniques5
       MR safety2
       Total464
      Tertiary screening
       Conference abstracts44
       MR-only review articles3
       Synthetic CT model development studies2
       Total49
      Abbreviations: CT = computed tomography; MR = magnetic resonance; MRI = magnetic resonance imaging; PET = positron emission tomography; UTE = ultrashort echo time.
      Included studies were organized into one or more of the following categories for review: (1) MR acquisition and synthetic CT generation verification, (2) MR distortion quantification and phantom development, (3) clinical validation of patient treatment positioning in an MR-only workflow, and (4) MR-only commissioning processes. These 4 categories will be discussed in more detail.

       MR acquisition and synthetic CT generation verification

      The systematic review identified 9 studies investigating MR acquisition or synthetic CT generation verification.
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Wyatt J.J.
      • Dowling J.A.
      • Kelly C.G.
      • et al.
      Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy.
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      A key summary of study results can be found in Table 2. All studies reported results relating to prostate cancer treatment planning. Studies reporting synthetic CT dosimetric accuracy findings were included when they were validating previously reported synthetic CT generation models as part of clinical implementation, rather than as part of the development of a synthetic CT model.
      Table 2Summary of the key results from the MR acquisition and synthetic CT generation verification studies
      “No. of patients in study” refers to the total number of patients recruited for MR-only investigations, “MR-only treated patients” refers to the number of patients planned and treated using the MR-only technique, and “sCT success rate” refers to the percentage of patients for whom a useable sCT was generated.
      AuthorYearsCT techniqueNo. patients in studyMR acquisitionsMR scanner and magnet strengthNo. of patients treated with MR onlysCT success rate (%)Other key information
      Christiansen et al
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      2017Philips - MRCAT30T1 mDIXONPhilips Ingenia, 1.5 T197 (29/30)
      • 1. MR synthetic CT generation failed in 1 case, reason unknown
      • 2. Dosimetric accuracy for gamma analysis of 2%/2 mm - median 100% in all structures
      • 3. Rectal gas found to be main contributor to dosimetric errors
      Kemppainen et al
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      2017Philips - MRCAT5T1 mDIXONPhilips Ingenia, 1.5 TXX
      • Only prostate patient data from study included
        • 1.
          Mean dosimetric accuracy (prostate patients) for 2%/2 mm and 1%/1 mm gamma analysis of 100% and 99.2%, respectively, within PTV
        • 2.
          Mean relative dose difference of 0.7% in PTV and <1.2% in OARs
      Maspero et al
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      2017Philips - MRCAT14T1 mDIXONPhilips Ingenia, 3 TXXMean relative dose difference between CT and sCT found to be 0.3% within the CTV and 0.04% within the whole body
      Persson et al
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      2017Spectronics Medical - MRIPlanner170T2 SPACEGE Discovery (3 T), GE Signa (3 T), Siemens Area (1.5 T),

      Siemens Skyra (3 T)
      X85 (145/170)
      • 1.
        Patient MR acquisition issues (no. of patients): distortion correction turned off (12), whole body not included in FoV (4), insufficient superior-inferior coverage (2), hip prosthesis patients (2), extreme rectum change between CT and MR (1)
      • 2.
        Mean dosimetric deviations of less than 0.3% for all targets and organs
      • 3.
        Multicentered (4 centers) study found insignificant differences found between range of treatment techniques, planning systems, prescribed doses, calculation models and target volumes
      Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      2017Philips - MRCAT48T1 mDIXONPhilips Ingenia, 3 T4287.5 (42/48)
      • 1.
        Patient MR acquisition issues (no. of patients): hip prosthesis patients (4), large patient exceeded MRCAT size limitations (2)
      • 2.
        Dedicated software used ed for contouring workflow. MR sequence blurring affected 2-dimensional DRR fiducial marker identification in 2 patient cases
      • 3.
        MRCAT failure modes: (i) presence of hip prosthesis, (ii) significant bone disease in pelvis, (iii) significant discrepancies from the bone model boundary conditions, and (iv) patient size exceeds 50 cm left-right or 30 cm anteroposterior
      • 4.
        Time saving of ∼15 minutes using MR-only simulation compared with CT-MR simulation, further 15-minute savings estimated in the future if logistic challenges resolved
      Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      2017Philips - MRCAT25T1 mDIXONPhilips Ingenia, 3 TXXMean relative dose difference between sCT and CT <0.5%
      Wyatt et al
      • Wyatt J.J.
      • Dowling J.A.
      • Kelly C.G.
      • et al.
      Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy.
      2017Dowling et al
      • Dowling J.A.
      • Sun J.
      • Pichler P.
      • et al.
      Automatic substitute computed tomography generation and contouring for magnetic resonance imaging (MRI)-alone external beam radiation therapy from standard MRI sequences.
      21T2 SPACEPhilips Magneto Espree, 1.5 TX54 (21/37)
      • Retrospective data collection
        • 1.
          Patient data set exclusions (no. of patients): required patient body outside of MR FoV (13), hip prostheses (2), gross patient motion (1)
        • 2.
          Dosimetric accuracy for 2%/2 mm gamma analysis: mean 98.9%, minimum 97.6%, and maximum 99.5% in all structures
      Kerkmeijer et al
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      2018Philips - MRCATNot knownT1 mDIXONPhilips (unknown)Yes, number unknownNot known
      • 1.
        Inclusion criteria: fiducial markers present in prostate
      • 2.
        Exclusion criteria: hip prostheses and contraindications for MRI
      Tenhunen et al
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      2018Korhonen et al
      • Korhonen J.
      • Kapanen M.
      • Keyriläinen J.
      • Seppälä T.
      • Tenhunen M.
      A dual model HU conversion from MRI intensity values within and outside of bone segment for MRI-based radiotherapy treatment planning of prostate cancer.
      250T1 mDIXONGE Optima, 1.5 T12592 (184/200)
      • 1.
        Patient MR acquisition issues (no. of patients): gold markers not identifiable (8), hip prosthesis related distortions (5), obesity (2), motion (1)
      • 2.
        CT vs MR-only patient treatment outcomes: PSA and acute toxicities results showed no significant differences between pathways
      • 3.
        Noted lack of support of MR-only workflow from technical software, including planning systems
      Abbreviations: CT = computed tomography; DRR = digitally reconstructed radiograph; MR = magnetic resonance; MRI = magnetic resonance imaging; PTV = planning target volume; sCT = synthetic computed tomography.
      “No. of patients in study” refers to the total number of patients recruited for MR-only investigations, “MR-only treated patients” refers to the number of patients planned and treated using the MR-only technique, and “sCT success rate” refers to the percentage of patients for whom a useable sCT was generated.
      Tenhunen et al,
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      Kerkmeijer et al,
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      Christiansen et al,
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      and Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      reported treating patients using an MR-only pathway, with the number of patients treated ranging from 125 to 1.
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      Tenhunen et al
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      also reported the MR-only patient cohort's initial clinical response in terms of early response PSA and acute toxicity follow-up.
      Persson et al,
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      Tenhunen et al,
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      and Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      reported their experiences of prospectively acquiring MR data for MR-only treatment planning in terms of their MR scan success rates and the issues that prevented successful scanning. In the case of Persson et al,
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      case, this was from a multicenter research-only study for commissioning purposes, whereas other experiences
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      resulted from treating their first MR-only patients. Wyatt et al
      • Wyatt J.J.
      • Dowling J.A.
      • Kelly C.G.
      • et al.
      Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy.
      reported MR scan success rates from retrospectively assessed MR data, whereas Christiansen et al
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      reported synthetic CT generation success rate but did not discuss issues regarding MR acquisition. In describing their clinical workflow for MR-only planning, Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      also reported a time savings when using an MR-only versus CT-MRI–based workflow.
      Previous studies
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Wyatt J.J.
      • Dowling J.A.
      • Kelly C.G.
      • et al.
      Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy.
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      reported validating the dosimetric accuracy of their respectively chosen synthetic CT solution in a clinical environment, as required for commissioning MR-only planning.

       MR distortion quantification and phantom development

      The systematic review identified 13 studies investigating MR distortion quantification methods or phantom development.
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      • Adjeiwaah M.
      • Bylund M.
      • Lundman J.A.
      • et al.
      Quantifying the effect of 3T magnetic resonance imaging residual system distortions and patient-induced susceptibility distortions on radiation therapy treatment planning for prostate cancer.
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      • Torfeh T.
      • Hammoud R.
      • Perkins G.
      • et al.
      Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
      • Price R.G.
      • Kadbi M.
      • Kim J.
      • et al.
      Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
      A key summary of study results can be found in Table 3.
      Table 3Summary of the key results from the MR distortion quantification and phantom development studies
      AuthorYearPhantom and softwarePhantom Shape (cm × cm × cm)MR scanner and magnet strengthOther key information
      Price et al
      • Price R.G.
      • Kadbi M.
      • Kim J.
      • et al.
      Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
      2015Philips temporal GNL phantom; in-house 3D distortion phantom; in-house software2D: 36 × 43 × 2

      3D: 46.5 × 35 × 16.8
      Philips Panorama, 1 T
      • 1.
        Gradient nonlinearity distortions found to be stable over 6-month period
      • 2.
        Vendor 3D distortion corrections maintained <1 mm distortion up to 9.5 cm from isocenter
      • 3.
        Postprocessing corrected distortions <1 mm for large FoV up to 25 cm from isocenter
      • 4.
        Significant inherent gradient nonlinearity distortions may be a specific feature of open-bore MR scanners, rather than cylindrical scanners, due to shorter gradient coils
      Sun et al
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      2015Self-developed pelvic-shape phantoms and software25 × 40 × 26Siemens Skyra, 3 T
      • 1.
        Phantom internal details: spherical and cylindrical inserts representing prostate, rectum bladder, and femoral heads based on average of 39 prostate patients or 11 plastic grid sheets
      • 2.
        Maximum distortion across phantom with 3D correction found to be 1.7 mm (75% quartile, 0.54 mm).
      • 3.
        Phantom end to end testing found mean dose difference of 1.1 cGy between CT and MR
      Walker et al
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      2015Self-developed large FoV phantom and softwareMax: 50 × 50 × 51.3Siemens Skyra, 3 T
      • 1.
        Maximum 3D distortion correction distortion was 4.08 mm for a 2-mm SE sequence.
      • 2.
        Within 152 mm of isocenter for 2 mm SE with 3D distortion correction, distortion ≤ 2 mm
      • 3.
        For the continuous moving-table mode, 1.1 mm/s was found to have the least distortion with a maximum of 4.4 mm and a distance of 140 mm within which the distortion was less than 2 mm.
      Huang 7et al
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      2016Self-developed large FoV phantom and software46.5 × 35 × 16.8Siemens Skyra, 3 TMean Bo distortion <1 mm found within a radius of 15 cm from the isocenter
      Torfeh et al
      • Torfeh T.
      • Hammoud R.
      • Perkins G.
      • et al.
      Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
      2016GE large-FoV phantom and in-house software50 × 50 × 50GE MR-Sim, 1.5 T
      • 1.
        In-house software validated with a mean distortion error of 0.15 mm
      • 2.
        Mean Bo distortion both in-plane and through plane found to be <2 mm within a radius of 25 cm when manufacturer 2D and 3D distortion applied as recommended
      • 3.
        Without distortion correction, the size of distortions made use for radiation therapy purposes unachievable
      Gustafsson et al
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      2017Spectronics large-FoV GRADE phantom and software50.2 × 40.4 × 53.4GE Discovery, 3 T
      • 1.
        Mean and maximum distortions <0.5 mm and <12.6 mm, respectively
      • 2.
        Maximum distortions: 0.43 mm at <100 mm, 0.82 mm at 100-150 mm, 1.85 mm at 150-200 mm, and 7.9 at 200-250 mm, increasing with radial distance from isocenter
      • 3.
        Structure deformation was minimal with mean magnitude 0.01 mm for internal structures and <0.33 mm for the full-body contour; mean percentage dose difference was ±0.02%.
      Kemppainen et al
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      2017Large-FoV phantom and software, unknown originMinimum:

      37.5 × 37.5 × 45.5
      Philips Ingenia, 1.5 T
      • 1.
        Mean system distortion of <1 mm measured within all PTVs with mean maximum distortion within patient body contours of <2 mm
      • 2.
        Effects of geometric distortion on dose calculation accuracy found to be <0.2% for all PTVs, with mean patient-induced distortions <1 mm in all cases
      Price et al
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      2017Self-developed large-FoV phantom in-house softwareMaximum:

      55 × 55 × 45
      Philips Panorama, 1 T; Philips Ingenia, 1.5 T; Philips Ingenia, 3 T
      • 1.
        Phantom modular to allow variation in setup for different scanners
      • 2.
        Setup reproducibility measured to be 0.1, 0, and –0.6 mm respectively in X, Y, and Z directions with negligible rotations
      • 3.
        Distortion <1 mm within 100 mm radially to isocenter
      Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      2017XXPhilips Ingenia, 3 TMean patient-induced susceptibility geometric distortion of –0.07 mm (range, –0.73 to –0.56 mm) and –0.2 mm (range: –0.62 to –0.35 mm) within the outer body and prostate, respectively
      Adjeiwaah et al
      • Adjeiwaah M.
      • Bylund M.
      • Lundman J.A.
      • et al.
      Quantifying the effect of 3T magnetic resonance imaging residual system distortions and patient-induced susceptibility distortions on radiation therapy treatment planning for prostate cancer.
      2018Spectronics phantom and software35.1 × 47 × 45.1GE Signa, 3 T
      • 1.
        For sequences of bandwidths of 122, 244, and 388 Hz, system distortions were <3.19 mm, <2.52 mm, and <2.08 mm within a radial distance of 25 cm from the isocenter, and the patient-induced distortions were <5.8 mm, <2.9 mm, and <1.5 mm, respectively
      • 2.
        Dosimetric analysis found that a mean dose difference of <0.5% was found between distorted and undistorted treatment plans
      • 3.
        Higher bandwidth sequences are recommended to minimize distortion effects
      Cunningham et al
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      2018Self-developed male pelvic-shape phantom23 × 38.1 × UnknownNot applicable
      • 1.
        External and internal organ shapes based on data from 19 prostate cancer patients
      • 2.
        Internal structure: pelvic bone anatomy, prostate, urethra, and fillable bladder and rectum
      • 3.
        Modular changes are possible to accommodate dosimetry inserts or organ changes
      • 4.
        Phantom able to accurately and reproducibly simulate rectum and bladder filling and to dosimetrically verify treatment plans, with an assessment plan found to have a dose difference of 1.5% between the calculated and measured doses
      Glide-Hurst et al
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      2018XXPhilips Panorama, 1 T; Philips Achieva, 1.5 T; Philips Ingenia, 3 T
      • 1.
        Empty, partially full, and full bladder states investigated over ∼45 minute scanning session
      • 2.
        Patient-induced susceptibility distortions were small with <2% of prostate and seminal vesicles voxels distorted by >0.5 mm and all-bladder voxels distorted by <1 mm.
      • 3.
        A significant change in rectal gas seen to increase distortion
      Wyatt et al
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      2018Spectronics Large FoV GRADE phantom and software50.2 × 40.4 × 53.4Siemens Magnetom Espree, 1.5 T; Siemens Prisma, 3 T; GE Signa PET-MR, 3 T
      • 1.
        Bo distortion measurements for intrascanning and interscanning sessions were repeatable
      • 2.
        Mean range of measurement for all scanners and sequences less than 1 mm, maximum ranges 2.9 mm and 2.6 mm for 1.5-T and 3-T scanners
      • 3.
        Phantom found to be relatively sensitive to large set up errors ∼1 mm translation or 1 degree of rotation
      Abbreviation: 2D = 2-dimensional; 3D = 3-dimensional; FoV = field of view; MR = magnetic resonance; PTV = planning target volume; SE = spin echo.
      Several authors
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      developed phantoms for use in measuring geometric distortions or end-to-end testing the MR-only pathway. Huang et al,
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      Price et al,
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      and Walker et al
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      developed large field of view (FoV) phantoms for assessing system (Bo) distortions and characterized their respective MR scanner distortions. Price et al
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      and Huang et al
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      assessed the setup reproducibility of their phantoms using CT scan testing methods, and Walker et al
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      assessed the effects of a continuous moving-table acquisition method on measured distortions with 0, 1.1, and 2 mm/s table speeds.
      Sun et al
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      and Cunningham et al
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      developed anthropomorphic pelvic-shaped phantoms for measuring system and patient-induced susceptibility distortions or end-to-end testing of the MR-only pathway. Both phantom designs were based on prostate patient anatomic sizes. The phantom of Sun et al
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      had 2 designs for end-to-end testing or geometric distortion assessment. Cunningham et al
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      designed the phantom so that it could simulate patient bladder and rectal filling for end-to-end testing, including dosimetric verification of treatment plans.
      Some authors
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      • Adjeiwaah M.
      • Bylund M.
      • Lundman J.A.
      • et al.
      Quantifying the effect of 3T magnetic resonance imaging residual system distortions and patient-induced susceptibility distortions on radiation therapy treatment planning for prostate cancer.
      investigated the effects of MR scanner distortions on patient treatments by applying measured or simulated distortions to patient treatment plans. Kemppainen et al
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      and Gustafsson et al
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      measured system-induced geometric distortions using large FoV phantoms for 15 and 10 patients, respectively. Tyagi et al,
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      Kemppainen et al,
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      and Glide-Hurst et al
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      measured patient-induced susceptibility geometric distortions for 20, 4, and 9 patients, respectively, with Glide-Hurst et al
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      assessing distortions for different patient bladder-filling states and scanner magnet strengths. Adjeiwaah et al
      • Adjeiwaah M.
      • Bylund M.
      • Lundman J.A.
      • et al.
      Quantifying the effect of 3T magnetic resonance imaging residual system distortions and patient-induced susceptibility distortions on radiation therapy treatment planning for prostate cancer.
      assessed the effects of MRI scanner-measured system and simulated patient-induced susceptibility distortions for 17 patients.
      Wyatt et al
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      evaluated the repeatability and setup sensitivity of the commercially available GRADE (Spectronics Medical AB, Helsingborg, Sweden), large-FoV distortion phantom. The distortion measurement repeatability was assessed for interscanning and intrascanning session variability. The setup sensitivity of the phantom was investigated by deliberately scanning the phantom with a 1-mm offset and 1 degree of rotation and assessing distortion variations.
      Torfeh et al
      • Torfeh T.
      • Hammoud R.
      • Perkins G.
      • et al.
      Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
      and Price et al
      • Price R.G.
      • Kadbi M.
      • Kim J.
      • et al.
      Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
      characterized their MR scanner system and gradient nonlinearity distortions respectively over large FoVs as required for MR-only planning. Torfeh et al
      • Torfeh T.
      • Hammoud R.
      • Perkins G.
      • et al.
      Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
      assessed the effects of manufacturer 2-dimensional and 3-dimensional distortion correction algorithms on clinically used radiation therapy sequences. Price et al
      • Price R.G.
      • Kadbi M.
      • Kim J.
      • et al.
      Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
      characterized and minimized inherent 2-dimensional and 3-dimensional large-FoV gradient nonlinearity distortions using postprocessing techniques.

       Clinical validation of patient treatment positioning in an MR-only workflow

      The systematic review identified 3 articles investigating the clinical validation of patient treatment positioning verification.
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      A key summary of study results can be found in Table 4. These studies have been included because they report patient treatment positioning verification results for previously reported MR-only treatment synthetic CT models as part of clinical implementation.
      Table 4Summary of the key results from the clinical validation of patient treatment positioning in MR-only workflow studies
      AuthorYearsCT techniqueNo. of PatientsDRR/CBCT2D methodCBCT methodInter/IntraobserverOther key information
      Korhonen et al
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      2015Korhonen et al
      • Korhonen J.
      • Kapanen M.
      • Keyriläinen J.
      • Seppälä T.
      • Tenhunen M.
      A dual model HU conversion from MRI intensity values within and outside of bone segment for MRI-based radiotherapy treatment planning of prostate cancer.
      DRR, 5; CBCT, 5DRR and CBCTManual, bony registrationAutomatic (bony and gray) value, 3D and 6D registrationDRR interobserver: 10CBCT (maximum difference)Gray value methodsCT vs CT—2 mm (3D), and 1.7 mm, 1.1° (6D)

      MR vs CT—4 mm (3D), and 3.5 mm, 1.6° (6D)
      Bone methodsCT vs CT—1.6 mm, and 1.3° (6D)
      DRRHeterogeneous sCT vs CT—manual registration errors were highest in the PA direction with mean differences of –0.3 ± 1 m and 0.3 ± 1.7 mm for kV and MV acquired positional images, respectively
      Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      2017Philips MR-CATDRR, 20; CBCT, 5DRR and CBCTManual, fiducial marker registrationManual, fiducial marker registrationXCBCT (mean difference)sCT vs CT: <1 ± 0.79 mm, <1 ± 0.89 mm, <0.5 ± 0.85 mm for LR, AP, and SI directions, respectively
      Other informationIndividual registration differences were observed up to 2 mm in some fractions with larger variations in prostate rotation
      DRR (mean difference)sCT vs CT: 0.3 mm, 0.3 mm, and 0.6 mm in the lateral, vertical and longitudinal directions, respectively
      Kemppainen et al
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      2018Philips MR-CAT20DRRManual, bony registrationXInterobserver, 5; intraobserver, 3DRR (mean difference)sCT vs CT: –0.5 mm, +0.1 mm and +0.1 mm in the vertical, longitudinal, and lateral directions, respectively
      Other information
      • 1.
        Repeatability coefficients were 2.1 mm vs 2.6 mm, 1.4 mm vs 2.1 mm, and 1.2 mm vs 1.4 mm in vertical, longitudinal, and lateral directions between CT and sCT, respectively
      • 2.
        Significant increase in intraobserver variability found for vertical, longitudinal, and lateral directions; however, magnitude was less than 0.5 mm in all directions
      • 3.
        MRCAT has positive effect on total geometric accuracy compared with 2-mm registration error of CT-MR pathway
      Abbreviations: 2D = 2-dimensional; 3D = 3-dimensional; 6D = 6 degrees of freedom; AP = anterior-posterior; CBCT = cone beam computed tomography; CT = computed tomography; DRR = digitally reconstructed radiograph; LR = left-right; MR = magnetic resonance; sCT = synthetic computed tomography; SI = superior-inferior.
      Tyagi et al,
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      Kemppainen et al,
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      and Korhonen et al
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      evaluated the accuracy of synthetic CTs as digitally reconstructed radiograph (DRR) reference images for treatment positional verification using orthogonal planar images,
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      cone beam CT (CBCT), or both.
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      and Kemppainen et al
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      investigated the Philips MRCAT synthetic CT solution. The DRR analysis by Kemppainen et al
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      included interobserver and intraobserver variability, separating the variability into systematic and random error contributions and comparing the total geometric accuracy to a reference of 2 mm error from CT to MR registration. The CBCT analysis by Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      was based on fiducial marker 3-dimensional CBCT scans, in which 5 CBCT scans were included for registration per patient.
      The DRR analysis by Korhonen
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      included interobserver variability and investigated the use of both the synthetic CT and MR images as reference images for CBCT registration. CBCT registrations were undertaken using ELEKTA x-ray volume imaging (XVI) software (Elekta, Stockholm, Sweden) for 5 patients, with 10 CBCT scans for each patient (50 CBCT registrations per reference modality).

       MR-only commissioning processes

      The systematic review identified 6 articles investigating MR-only commissioning processes.
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      • Kapanen M.
      • Collan J.
      • Beule A.
      • et al.
      Commissioning of MRI-only based treatment planning procedure for external beam radiotherapy of prostate.
      • Kim J.
      • Miller B.
      • Siddiqui M.S.
      • et al.
      FMEA of MR-only treatment planning in the pelvis.
      • Korsholm M.E.
      • Waring L.W.
      • Edmund J.M.
      A criterion for the reliable use of MRI-only radiotherapy.
      • Palmer E.
      • Persson E.
      • Ambolt P.
      • et al.
      Cone beam CT for QA of synthetic CT in MRI only for prostate patients.
      A key summary of study results can be found in Table 5.
      Table 5Summary of the key results from the MR-only commissioning processes studies
      AuthorYearNo. of patients in study
      ”No. patients in Study” refers to the total number of patients recruited for the MR-only investigations.
      Other key information
      Kapanen et al
      • Kapanen M.
      • Collan J.
      • Beule A.
      • et al.
      Commissioning of MRI-only based treatment planning procedure for external beam radiotherapy of prostate.
      2013XProposed calibration and testing procedures for verification of the treatment isocenter position, geometric accuracy, and other basic QA with an ACR phantom
      Korsholm et al
      • Korsholm M.E.
      • Waring L.W.
      • Edmund J.M.
      A criterion for the reliable use of MRI-only radiotherapy.
      201421A statistical model approach to assessing the accuracy of sCT calculation was used where the criteria of accuracy was considered to be 95% of patients having an uncertainty in dose calculation within the PTV within 2% of the CT dose
      Maspero et al
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      201714For electron density conversion, sCT generation, and interscan difference, average dose difference in the CTV of 0.7% ± 0.2%, 0.16% ± 0.13%, and 0.01% ± 0.35% and in the whole body of 0.1% ± 0.03%, –0.03% ± 0.02%, and 0 ± 0.06% were found, respectively
      Kerkmeijer et al
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      2018XRecommended requirements for MR-only radiation therapy clinical implementation including geometric accuracy, treatment position MR acquisition, sCT generation, MRI-based OAR delineation, and protocol optimization and MRI-based treatment position verification
      Kim et al
      • Kim J.
      • Miller B.
      • Siddiqui M.S.
      • et al.
      FMEA of MR-only treatment planning in the pelvis.
      2018X
      • 1.
        Many processes and therefore failure modes are shared between CT-MR and MR-only workflows with the highest failure modes related to changes in target location due to internal anatomy changes, in these cases current mitigation processes were still valid
      • 2.
        The highest risk failure modes for the MR-only workflow alone related to the sCT generation process, including: inaccuracies in target delineation on MR images, insufficient management of patient- and system-level distortions and inaccurate bone volumes
      • 3.
        Mitigation strategies for failures include sufficient staff training and a robust quality-control and quality-assurance program
      Palmer et al
      • Palmer E.
      • Persson E.
      • Ambolt P.
      • et al.
      Cone beam CT for QA of synthetic CT in MRI only for prostate patients.
      201810
      • 1.
        The CBCT system was stable over time in HU (standard deviation <40 HU) and the variation in HU between CT and CBCT was found to be minimal (<60 HU)
      • 2.
        A comparison of the dose distributions between sCT and CT compared with sCT and CBCT found mean dose differences for all metrics of ≤1%
      • 3.
        The CBCT system can be considered to be similar to a CT system and can be used as a clinically feasible QA procedure
      Abbreviations: ACR = American College of Radiology; CBCT = cone beam CT; CT = computed tomography; HU = Hounsfield units; MR = magnetic resonance; MRI = magnetic resonance imaging; PTV = planning target volume; QA = quality assurance; sCT = synthetic CT.
      ”No. patients in Study” refers to the total number of patients recruited for the MR-only investigations.
      Kerkmeijer et al,
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      Kapanen et al,
      • Kapanen M.
      • Collan J.
      • Beule A.
      • et al.
      Commissioning of MRI-only based treatment planning procedure for external beam radiotherapy of prostate.
      and Kim et al
      • Kim J.
      • Miller B.
      • Siddiqui M.S.
      • et al.
      FMEA of MR-only treatment planning in the pelvis.
      reported experiences related to commissioning an MR-only pathway. Kerkmeijer et al
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      and Kapanen et al
      • Kapanen M.
      • Collan J.
      • Beule A.
      • et al.
      Commissioning of MRI-only based treatment planning procedure for external beam radiotherapy of prostate.
      used their experiences of commissioning an MR-only pathway and an MR simulator, respectively, to present recommendations for clinically commissioning an MR-only pathway, including proposing quality-assurance testing and associated levels of acceptability with individual pathway components. Kim et al
      • Kim J.
      • Miller B.
      • Siddiqui M.S.
      • et al.
      FMEA of MR-only treatment planning in the pelvis.
      used a failure mode and effects analysis methodology to systematically assess the risks—and their frequency, severity, and detectability—of an MR-only planning pathway compared with CT-MR based pathway. This analysis included mapping the respective elements required for an MR-only pathway, their risks, and associated mitigation strategies.
      Maspero et al
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      and Korsholm et al
      • Korsholm M.E.
      • Waring L.W.
      • Edmund J.M.
      A criterion for the reliable use of MRI-only radiotherapy.
      reported synthetic CT accuracy assessment methodologies. Maspero et al
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      quantified the confounding factors in MR-only dose calculation accuracy assessments for patients with prostate cancer, including interscan differences (setup and positioning differences, MR-related geometric inaccuracy, and registration errors) and synthetic CT generation and electron density conversion errors. Korsholm et al
      • Korsholm M.E.
      • Waring L.W.
      • Edmund J.M.
      A criterion for the reliable use of MRI-only radiotherapy.
      developed a statistical approach to evaluating the significance of errors introduced by MR-only planning compared with CT-based planning, with the criterion that 95% of patients should have an uncertainty in dose calculation within 2% of the CT dose for relevant structures.
      Palmer et al
      • Palmer E.
      • Persson E.
      • Ambolt P.
      • et al.
      Cone beam CT for QA of synthetic CT in MRI only for prostate patients.
      developed and validated a quality assurance procedure for assessing synthetic CT clinical feasibility using kV-CBCT, where CBCT scans were used to recalculate the synthetic CT treatment plan as a check of its dose calculation accuracy.

      Discussion

      A wide range of findings are reported in this systematic review. Several key findings are seen in the literature, and these are highlighted here before being discussed in more detail later. These findings are (1) MR-only planning has been clinically implemented for prostate cancer treatments; (2) a substantial amount of work remains to translate MR-only planning into widespread clinical implementation for all pelvic sites; (3) MR scanner distortions are no longer a barrier to MR-only planning, but they must be managed appropriately; (4) MR-only based patient positioning verification shows promise, but limited evidence is reported in the literature and further investigation is required; and (5) a number of MR-only commissioning processes have been reported, which can aid centers as they undertake local commissioning, but this needs to be formalized in guidance from national bodies.
      As highlighted, in 4 studies patients with prostate cancer were treated using an MR-only planning solution, showing that clinical implementation is achievable.
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      It is interesting to note that all commissioning work identified in this review was also focused on prostate treatments. This is a natural starting point for pelvic MR-only planning because other pelvic sites (rectum, bladder, anus, gynecological) have a number of additional challenges associated with them, including differences in male and female anatomy, significantly larger treatment volumes, and non–fiducial-marker-based 3-dimensional imaging requirements, which makes their implementation more complex. It is important to note that the majority of work discussed is translatable to other cancer sites; however, it is clear that a significant amount of work remains to widen the implementation of MR-only planning to all pelvic cancer treatments.
      This review identified 3 key areas that were investigated for clinical implementation purposes: MR acquisition and synthetic CT generation verification, MR distortion quantification and phantom development, and clinical validation of patient treatment positioning in an MR-only workflow. In each, no major barriers to implementation were found, and a number of publications reported commissioning methodologies that will benefit the wider community by providing guidance for local centers to use within their own MR-only clinical commissioning.
      The first step to implementing an MR-only pathway is ensuring that sufficient and suitable MR data acquisition is achieved. A high success rate of acquiring MRI that is usable for synthetic CT generation is key to the widespread implementation of MR-only techniques because this will limit the need for additional CT scans. Persson et al,
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      Tenhunen et al,
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      and Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      all described their success rates in prospective studies and categorized the identified issues related to scanning. The similar scan success rates (85%-92%) suggest that this is an achievable percentage in any center, particularly considering the multicenter study by Persson et al.
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      The differences in success rate can be explained by the variations in study design. The exclusion criteria of Persson et al
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      included patients with hip prostheses and operator error as valid reasons for an unsuccessful MRI scan, whereas Tenhunen et al
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      and Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      had no exclusion criteria. Wyatt et al
      • Wyatt J.J.
      • Dowling J.A.
      • Kelly C.G.
      • et al.
      Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy.
      also analyzed their successful scanning rate (54%); however, their rate was severely affected by a lack of scanning guidance for MR operators caused by the retrospective design of the study. However, the study does still provide useful information regarding common issues with MR acquisition in this context. Centers should ensure that training from experienced personnel is provided for MR scan operators and consider methods to identify errors at the point of acquisition to ensure MR scan success. Tenhunen et al
      • Tenhunen M.
      • Korhonen J.
      • Kapanen M.
      • et al.
      MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
      and Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      identified several patient and hardware- and software-related issues that also prevented successful MR-only planning and therefore required a percentage of patients to revert to a CT-MRI–based pathway. Although further development of MR-only solutions could lead to a reduction in patients requiring an additional CT scan, provisions should still be made for CT-based planning to occur. In addition, these studies do not discuss patients who have contraindications to MRI and therefore will always require a CT-only pathway.
      Christiansen et al,
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      did not report MR acquisition success, but they described their synthetic CT generation failure rate, finding that 3% (1 of 30) of synthetic CT examinations failed to generate using the Philips MRCAT solution. This was considered to be due to the software's “sanity” check ability to prevent obviously erroneous synthetic CT generation, although the exact cause was not established. This finding highlights that synthetic CT generation methods require input data to follow clearly defined criteria to be successful, and it is a beneficial feature that the Philips MRCAT safeguards against inappropriate data, defined as including large patient sizes, large disease sites (300 mm or greater scan lengths), and hip prostheses.
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      It is of note that Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zelefsky M.M.
      • et al.
      Clinical workflow for MR-only simulation and planning in prostate.
      did not have any similar issues. This result could have been due to a systematic difference in pathway, such as Tyagi's use of a specific mold for each patient to achieve a more robust patient position, or a non-systematic, patient-specific, issue. This is another example of the variety of errors associated with an MR-only planning pathway that require careful assessment.
      For acquired MR data to be clinical usable, their dosimetric accuracy needs to be quantified robustly as within acceptable limits. Dosimetric accuracy of prostate synthetic CT solutions was investigated by the majority of studies and considered to be clinically acceptable in all cases.
      • Persson E.
      • Gustafsson C.
      • Nordström F.
      • et al.
      MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
      • Christiansen R.L.
      • Jensen H.R.
      • Brink C.
      Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Wyatt J.J.
      • Dowling J.A.
      • Kelly C.G.
      • et al.
      Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy.
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      The similar results of these studies, despite significant differences in study design including various synthetic CT generation methods, shows that the dosimetric accuracy of synthetic CT techniques is broadly reproducible across a wide range of clinical systems and techniques, including multiple commercial options available for prostates.
      • Siversson C.
      • Nordström F.
      • Nilsson T.
      • et al.
      Technical note: MRI only prostate radiotherapy planning using the statistical decomposition algorithm.
      • Köhler M.
      • Vaara T.
      • Van Grootel M.
      • et al.
      MR-Only Simulation for Radiotherapy Planning [white paper].
      The presented studies also provide a suitable blueprint for centers wishing to begin clinical implementation of MR-only planning themselves regarding dosimetric accuracy assessment. These studies progress by first assessing dosimetric accuracy through research-only studies, followed by end-to end pathway testing and eventual implementation only when the local results provided sufficient confidence that the MR-only planning was sufficiently dosimetrically accurate to be used without CT for assurance.
      In addition to dosimetric accuracy, another key criterion of useable MR data for MR-only treatment planning is that it be geometrically accurate. To be sufficiently confident of this for clinical implementation requires robust quality assurance techniques, phantoms, and the characterization of the MRI distortions. The reported studies focused on designing suitable phantoms for measuring geometric distortions or end-to-end testing the MR-only pathway,
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      the quantification of distortions on patient data,
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      • Adjeiwaah M.
      • Bylund M.
      • Lundman J.A.
      • et al.
      Quantifying the effect of 3T magnetic resonance imaging residual system distortions and patient-induced susceptibility distortions on radiation therapy treatment planning for prostate cancer.
      or the reproducibility of distortion measurements
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      and provide information to aid distortion commissioning for an MR-only pathway.
      Distortions within 1-T,
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      • Price R.G.
      • Kadbi M.
      • Kim J.
      • et al.
      Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
      1.5-T,
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      • Torfeh T.
      • Hammoud R.
      • Perkins G.
      • et al.
      Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
      and 3-T scanners
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      from a range of manufacturers, including Siemens,
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      Philips,
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      • Price R.G.
      • Kadbi M.
      • Kim J.
      • et al.
      Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
      and GE,
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      • Torfeh T.
      • Hammoud R.
      • Perkins G.
      • et al.
      Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
      were measured within a satisfactory range for MR-only planning, considered to be 2 mm.
      • Weygand J.
      • Fuller C.D.
      • Ibbott G.S.
      • et al.
      Spatial precision in magnetic resonance imaging-guided radiation therapy: The role of geometric distortion.
      Unlike the other presented studies, Wyatt et al
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      and Price et al
      • Price R.G.
      • Kadbi M.
      • Kim J.
      • et al.
      Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
      suggested that the majority of distortions measured as part of their studies were larger than clinically acceptable for MR-only planning. However, both authors noted that their MR sequences were not optimized for clinical use because their acquired sequence bandwidths were insufficient to reduce distortion to within acceptable levels, where a minimum suitable bandwidth is considered to be 220 Hz/pixel at 1.5 T and 440 Hz/pixel at 3 T. This is a timely reminder that scanners and clinically used scans require distortion characterization to ensure they are suitable for use.
      It is important that distortions be placed in context by evaluating their effect on patient treatments. A number of studies did this by assessing the effects of different distortions when applied to patient treatment plans, with Kemppainen et al
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      and Gustafsson et al
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      assessing system distortions; Tyagi et al,
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      Kemppainen et al,
      • Kemppainen R.
      • Suilamo S.
      • Tuokkola T.
      • et al.
      Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
      and Glide-Hurst
      • Glide-Hurst C.
      • Nejad-Davarani S.
      • Weiss S.
      • et al.
      Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
      assessing patient-induced susceptibility distortions; and Adjeiwaah et al
      • Adjeiwaah M.
      • Bylund M.
      • Lundman J.A.
      • et al.
      Quantifying the effect of 3T magnetic resonance imaging residual system distortions and patient-induced susceptibility distortions on radiation therapy treatment planning for prostate cancer.
      assessing both system and patient-induced susceptibility distortions. These results broadly showed that these distortions can be considered to have negligible effects on the patient plan, although their assessment and subsequent protocol optimization is vital. This is particularly true when regarding patient-induced distortions, which cannot be corrected systematically because they vary between patients; however, these studies provide confidence that their effects can be quantified or negated for a range of scanners and magnet strengths. Similar investigations should form part of any center's clinical implementation of MR-only planning to allow local distortion effects to be quantified and assessed as clinically significant or not. It is noted these studies were undertaken with low patient numbers (20 or fewer), and in none of the studies were distortions correlated with patient size. Because distortions increase with distance from the isocenter, their effects will increase with patient size. Without information relating to patient sizes, it is not possible to assess whether the true effect of the distortions on larger patients has been quantified. Selection of a large range of patient sizes and quantification of the impact of distortion as a function of patient size potentially would be of more value to a commissioning center than an attempt to establish the “average” patient size.
      It is also to the wider radiation therapy community's benefit that self-developed phantom designs be detailed in the literature
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      to allow centers to replicate these phantoms. It was noticeable that for pelvic MR-only large-FoV distortion measurements, these studies used only 2 commercially available phantoms—the Spectronics Medical's GRADE
      • Gustafsson C.
      • Nordström F.
      • Persson E.
      • Brynolfsson J.
      • Olsson L.E.
      Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
      • Wyatt J.
      • Hedley S.
      • Johnstone E.
      • et al.
      Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
      and GE's large-FoV phantoms
      • Torfeh T.
      • Hammoud R.
      • Perkins G.
      • et al.
      Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
      —with the remaining studies developing their own phantoms.
      • Price R.G.
      • Knight R.A.
      • Hwang K.P.
      • et al.
      Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
      • Huang K.C.
      • et al.
      Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
      • Walker A.
      • Cao Y.
      • Baharom U.
      • Balter J.M.
      Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      However, it should be noted that there are several other commercially available phantoms that have not been reported here, including the Quasar MRID 3D, CIRS large FoV, Phantomlab MagPhan RT, and Philips MRI distortion phantoms. It is possible that the use of these phantoms is not reflected in this systematic review because of the search criteria focusing on MR-only clinical implementation pathways, rather than CT-MRI pathways. As such, further comment on their potential benefit in this context is not possible.
      The anthropomorphic phantoms from Sun et al
      • Sun J.
      • Dowling J.
      • Pichler P.
      • et al.
      MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
      and Cunningham et al
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      are a beneficial development in phantom design because they allow quantitative end-to-end testing, including dose measurements within the phantoms. Of interest is the ability of one set of phantoms
      • Cunningham J.M.
      • Barberi E.A.
      • Miller J.
      • et al.
      Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
      to model physiologic changes in the bladder and rectum, which will improve the commissioning process by allowing the effects of patient anatomic changes to be assessed in a quantitative and reproducible manner. Further development of anthropomorphic phantoms could increase their use in the quality assurance of MR-only planning as it develops as a clinical technique.
      For MR-only planning to be implemented, MR data are required to be used for patient position verification purposes before treatment. Publications relating to data assessment as part of clinical implementation were limited, however, with only Tyagi et al,
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      Kemppainen et al,
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      and Korhonen et al
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      assessing prostate patient treatment positioning accuracy. All 3 studies assessed DRR positioning, with results showing broad agreement between the use of planning CT and synthetic CT–generated DRRs, providing confidence that synthetic CT–generated DRRs can produce clinically acceptable results. Korhonen et al
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      and Kemppainen et al
      • Kemppainen R.
      • Vaara T.
      • Joensuu T.
      • Kiljunen T.
      Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
      also investigated interobserver and intraobserver variability for DRR registrations and found it to be clinically insignificant.
      Three-dimensional CBCT–based patient treatment positioning was investigated using manual fiducial marker registration
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      and the automatic bone and gray-value registration methods of Elekta's XVI system.
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      Synthetic CT–to-CBCT registrations were comparable to planning CT-to-CBCT registrations (mean differences <1 mm), indicating that synthetic CT data sets can replace CT data sets for manual or automatic registrations and for patient treatment positioning. Tyagi et al
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      also noted that, anecdotally, clinicians were happy with the delineations of bladder and rectum on the synthetic CT. It is interesting that the synthetic CT–to-CT results varied between Korhonen et al
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      and Tyagi et al.
      • Tyagi N.
      • Fontenla S.
      • Zhang J.
      • et al.
      Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
      This could be influenced by a number of factors, including the difference in matching technique (automatic vs manual) or the inclusion of a patient mold within Tyagi's study to improve setup reproducibility.
      MR-to-CBCT registrations were also assessed by Korhonen et al,
      • Korhonen J.
      • Kapanen M.
      • Sonke J.J.
      • et al.
      Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
      but they did not replicate the same level of similarity to planning CT-to-CBCT registrations as synthetic CT–to-CBCT registrations. This result is understandable because XVI uses a chamfer matching algorithm and is optimized for registering data sets of the same modality (ie, CT to CBCT), and registrations could improve if more suitable mutual information algorithms were used. It is an indicator that CT cannot be simply replaced with MRI and that further commercial support and investment in this field is required. In addition, it is important to note that although differences are seen between CT and MR registrations, from the data shown, neither CT nor MR can be determined as more accurate because there is no ground truth for comparison. It can only be determined that the registrations produce different results. To resolve which modality is more accurate, manual landmark evaluation can be used for an initial comparison, and a future potential solution would be to use an anthropomorphic phantom that could provide the required ground truth information. It can be hypothesized that it would be best to register MR to CBCT, rather than synthetic CT to CBCT, because this would mean that real rather than synthesized data were being used, thus theoretically improving the registration accuracy. These findings suggest that MR-only pathways exist that allow reproducible patient positioning verification to be completed. In addition, these studies provide a suitable methodology for a center looking to implement MR-only planning with respect to the assessment of patient treatment positioning accuracy and reproducibility.
      A wide variety of processes and experiences relating to commissioning an MR-only pathway were reported for prostate treatments. The experiences of Kerkmeijer et al
      • Kerkmeijer L.G.W.
      • Maspero M.
      • Meijer G.J.
      • et al.
      Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
      and Kapanen et al
      • Kapanen M.
      • Collan J.
      • Beule A.
      • et al.
      Commissioning of MRI-only based treatment planning procedure for external beam radiotherapy of prostate.
      in terms of workflow, equipment, and commissioning requirements provide substantial amounts of information, which is particularly beneficial because these processes, within this early phase of clinical implementation, are not well established. It is a challenging process to determine the commissioning and routine workflow to ensure optimal performance and the highest quality of patient care; therefore, more publications detailing individual centers’ experiences, such as these, would be welcome until this technique is more firmly embedded in routine clinical practice or guidance documents are published. The information provided within the work of Kim et al
      • Kim J.
      • Miller B.
      • Siddiqui M.S.
      • et al.
      FMEA of MR-only treatment planning in the pelvis.
      provides useful tools for identifying risks and highlights many risks that will be shared among all MR-only pathways; it also suggests mitigation strategies to lessen their influence. That the greatest source of risk is the synthetic CT generation process is not a surprising result; however, the strength of this methodology is that it provides an overall framework for assessing, comparing, and minimizing risks. This methodology also allows the user to have confidence that their MR-only pathway is optimized to protect from errors as much as is reasonably practicable. In addition, as a process, it can be repeated over time to reassess and fine-tune the pathway continually; it is also applicable to any future MR-only treatment sites in addition to prostate cancer, which was presented here.
      The quantification of the accuracy of a center's local synthetic CT generation technique is a key stage of commissioning an MR-only pathway, and the studies by Maspero et al
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      and Korsholm et al
      • Korsholm M.E.
      • Waring L.W.
      • Edmund J.M.
      A criterion for the reliable use of MRI-only radiotherapy.
      provide differing methods of undertaking this. It is interesting to see that Maspero et al
      • Maspero M.
      • Seevinck P.R.
      • Schubert G.
      • et al.
      Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
      found that electron density conversion (from CT scan–generated Hounsfield units to electron density via an election density plotted curve) was the greatest confounding factor, followed by synthetic CT generation (the assigning of Hounsfield units to MR scan voxels to produce the synthetic CT). Interscan differences (setup and positioning differences between CT and MRI scans, MR geometric inaccuracy, and CT-MRI registration errors as required for comparison) produced almost a negligible difference in result. This finding suggests that the commissioning process should also focus on appropriate electron density curve calibration as a key part of the commissioning process.
      Palmer et al
      • Palmer E.
      • Persson E.
      • Ambolt P.
      • et al.
      Cone beam CT for QA of synthetic CT in MRI only for prostate patients.
      present a method of validating synthetic CT generations using collected patient CBCT data and provide a tool by which commissioning centers can ensure further confidence over the accuracy of their treatment planning pathway. As previously discussed by Kim et al,
      • Kim J.
      • Miller B.
      • Siddiqui M.S.
      • et al.
      FMEA of MR-only treatment planning in the pelvis.
      the generation of synthetic CTs is a major risk in the MR-only pathway, and the challenge of ensuring robust patient treatment on an individual basis is nontrivial. The method of Palmer et al would directly allow a gross error check on the treatment plan, which could highlight potential issues at the beginning of a patient's treatment. Palmer et al
      • Palmer E.
      • Persson E.
      • Ambolt P.
      • et al.
      Cone beam CT for QA of synthetic CT in MRI only for prostate patients.
      noted, however, that further analysis of this technique would be beneficial, as only simple errors were assessed within the validation presented; as a consequence, its sensitivity to less-noticeable errors is uncertain.
      The studies identified here are a significant step toward widespread pelvic MR-only clinical implementation; however, further attention is required in several areas. As discussed previously, the translation of this technique to other clinical pelvic sites is a significant challenge that should not be underestimated. Several studies reported issues associated with processing data within MR-only pathways. These issues may be due to a lack of support for the MR-only workflow by radiation therapy vendors. Clinical treatment planning software and further collaboration, investment, or support from commercial companies would be beneficial. Unseen in the literature was a long timescale (≥1 year) Bo distortion evaluation study focusing on its effects on an MR-only pathway. Long timescale changes in Bo distortion could have a significant influence on resultant MRI geometric accuracy, which would require correction to prevent the translation of errors into the planning process. Such a study would demonstrate the reliability and reproducibility of scanner Bo distortion over time and therefore provide evidence for distortion quality assurance frequency recommendations. Patient treatment positioning verification within clinical implementation was addressed in only 3 studies. Although the results presented were encouraging and suggested that MR-only techniques can accurately be used for patient treatment positioning, there is plenty of evidence yet to be gathered. All these studies involved small cohorts of patients and used the synthetic CT generation method of the Philips prostate MRCAT
      • Köhler M.
      • Vaara T.
      • Van Grootel M.
      • et al.
      MR-Only Simulation for Radiotherapy Planning [white paper].
      or similar,
      • Korhonen J.
      • Kapanen M.
      • Keyriläinen J.
      • Seppälä T.
      • Tenhunen M.
      A dual model HU conversion from MRI intensity values within and outside of bone segment for MRI-based radiotherapy treatment planning of prostate cancer.
      which are not comparable to all methods of synthetic CT generation. In addition, the majority of results were collected with manual registration techniques, whereas it is common in the clinic to use a manufacturer's automated or semiautomated technique. The effects different clinical equipment and techniques, in larger patient cohorts, need to be investigated fully.
      Despite the variety of publications related to MR-only commissioning and individual centers experiences, the radiation therapy community is so varied in term of equipment, resources, and technique that there is significant scope for further experience to be reported in the literature and consensus guidelines to be produced by early adopters and national bodies. There is also a substantial need for more studies to begin providing evidence of the benefit of using MR-only planning, such as improved patient outcomes or treatment pathway improvements.

      Conclusion

      MR-only planning has been clinically implemented for the treatment of prostate cancer; however, further research is required to develop MR-only planning for other pelvic sites. In particular, the accuracy of synthetic CT generation models for female anatomy requires further reporting within the literature. MR scanner distortions are no longer a barrier to MR-only planning, although they must be managed appropriately, whereas MR data acquisition and synthetic CT generation for prostate treatments have been shown to be sufficiently accurate for clinical use. The clinical implementation of MR-only patient treatment positioning verification remains underreported in the literature and requires substantial investigation to allow its widespread use. The range of investigations reported here are a suitable starting point for radiation therapy centers aiming to clinically implement MR-only planning; however, further evidence and regulation is required, including the publication of consensus guidelines from early adopters and governing bodies.

      Supplementary Data

      References

        • Njeh C.F.
        Tumor delineation: The weakest link in the search for accuracy in radiotherapy.
        J Med Phys. 2008; 33: 136-140
        • Sander L.
        • Langkilde N.C.
        • Holmberg M.
        • Carl J.
        MRI target delineation may reduce long-term toxicity after prostate radiotherapy.
        Acta Oncol. 2014; 53: 809-814
        • Wortel R.C.
        • Heemsbergen W.D.
        • Smeenk R.J.
        • et al.
        Local Protocol Variations for Image Guided Radiation Therapy in the Multicenter Dutch Hypofractionation (HYPRO) Trial: Impact of rectal balloon and MRI delineation on anorectal dose and gastrointestinal toxicity levels.
        Int J Radiat Oncol Biol Phys. 2017; 99: 1243-1252
        • Johnstone E.
        • Wyatt J.J.
        • Henry A.M.
        • et al.
        Systematic review of synthetic computed tomography generation methodologies for use in magnetic resonance imaging-only radiation therapy.
        Int J Radiat Oncol Biol Phys. 2018; 100: 199-217
        • Owrangi A.M.
        • Greer P.B.
        • Glide-Hurst C.K.
        MRI-only treatment planning: Benefits and challenges.
        Phys Med Biol. 2018; 63: 05TR01
        • Persson E.
        • Gustafsson C.
        • Nordström F.
        • et al.
        MR-OPERA: A multicenter/multivendor validation of magnetic resonance imaging-only prostate treatment planning using synthetic computed tomography images.
        Int J Radiat Oncol Biol Phys. 2017; 99: 692-700
        • Karlsson M.
        • Karlsson M.G.
        • Nyholm T.
        • Amies C.
        • Zackrisson B.
        Dedicated magnetic resonance imaging in the radiotherapy clinic.
        Int J Radiat Oncol Biol Phys. 2009; 74: 644-651
        • Weygand J.
        • Fuller C.D.
        • Ibbott G.S.
        • et al.
        Spatial precision in magnetic resonance imaging-guided radiation therapy: The role of geometric distortion.
        Int J Radiat Oncol Biol Phys. 2016; 95: 1304-1316
        • Edmund J.M.
        • Nyholm T.
        A review of substitute CT generation for MRI-only radiation therapy.
        Radiat Oncol. 2017; 12 ([Review]): 26
        • Siversson C.
        • Nordström F.
        • Nilsson T.
        • et al.
        Technical note: MRI only prostate radiotherapy planning using the statistical decomposition algorithm.
        Med Phys. 2015; 42: 6090-6097
        • Köhler M.
        • Vaara T.
        • Van Grootel M.
        • et al.
        MR-Only Simulation for Radiotherapy Planning [white paper].
        Koninklijke Philips, Amsterdam2015
        • Philips Healthcare
        MR-only simulation pelvis 2018.
        (Available at:)
        • Moher D.
        • Liberati A.
        • Tetzlaff J.
        • Altman D.G.
        • PRISMA Group
        Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement.
        PLoS Med. 2009; 6: e1000097
        • Tenhunen M.
        • Korhonen J.
        • Kapanen M.
        • et al.
        MRI-only based radiation therapy of prostate cancer: Workflow and early clinical experience.
        Acta Oncol. 2018; 57: 902-907
        • Kerkmeijer L.G.W.
        • Maspero M.
        • Meijer G.J.
        • et al.
        Magnetic resonance imaging only workflow for radiotherapy simulation and planning in prostate cancer.
        Clin Oncol. 2018; 30: 692-701
        • Christiansen R.L.
        • Jensen H.R.
        • Brink C.
        Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer.
        Acta Oncol. 2017; 56: 787-791
        • Tyagi N.
        • Fontenla S.
        • Zelefsky M.M.
        • et al.
        Clinical workflow for MR-only simulation and planning in prostate.
        Radiat Oncol. 2017; 12: 119
        • Tyagi N.
        • Fontenla S.
        • Zhang J.
        • et al.
        Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis.
        Phys Med Biol. 2017; 62: 2961-2975
        • Kemppainen R.
        • Suilamo S.
        • Tuokkola T.
        • et al.
        Magnetic resonance-only simulation and dose calculation in external beam radiation therapy: A feasibility study for pelvic cancers.
        Acta Oncol. 2017; 56: 792-798
        • Wyatt J.J.
        • Dowling J.A.
        • Kelly C.G.
        • et al.
        Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy.
        Phys Med Biol. 2017; 62: N548-N560
        • Maspero M.
        • Seevinck P.R.
        • Schubert G.
        • et al.
        Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT.
        Phys Med Biol. 2017; 62: 948-965
        • Price R.G.
        • Knight R.A.
        • Hwang K.P.
        • et al.
        Optimization of a novel large field of view distortion phantom for MR-only treatment planning.
        J Appl Clin Med Phys. 2017; 18: 51-61
        • Huang K.C.
        • et al.
        Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology.
        Phys Med Biol. 2016; 61: 774-790
        • Walker A.
        • Cao Y.
        • Baharom U.
        • Balter J.M.
        Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom.
        Med Phys. 2015; 42: 1982-1991
        • Sun J.
        • Dowling J.
        • Pichler P.
        • et al.
        MRI simulation: End-to-end testing for prostate radiation therapy using geometric pelvic MRI phantoms.
        Phys Med Biol. 2015; 60: 3097-3109
        • Cunningham J.M.
        • Barberi E.A.
        • Miller J.
        • et al.
        Development and evaluation of a novel MR-compatible pelvic end-to-end phantom.
        J Appl Clin Med Phys. 2019; 20: 265-275
        • Gustafsson C.
        • Nordström F.
        • Persson E.
        • Brynolfsson J.
        • Olsson L.E.
        Assessment of dosimetric impact of system specific geometric distortion in an MRI only based radiotherapy workflow for prostate.
        Phys Med Biol. 2017; 62: 2976-2989
        • Glide-Hurst C.
        • Nejad-Davarani S.
        • Weiss S.
        • et al.
        Per-organ assessment of subject-induced susceptibility distortion for MR-only male pelvis treatment planning.
        Radiat Oncol. 2018; 13: 149
        • Adjeiwaah M.
        • Bylund M.
        • Lundman J.A.
        • et al.
        Quantifying the effect of 3T magnetic resonance imaging residual system distortions and patient-induced susceptibility distortions on radiation therapy treatment planning for prostate cancer.
        Int J Radiat Oncol Biol Phys. 2018; 100: 317-324
        • Wyatt J.
        • Hedley S.
        • Johnstone E.
        • et al.
        Evaluating the repeatability and set-up sensitivity of a large field of view distortion phantom and software for magnetic resonance-only radiotherapy.
        Phys Imaging Radiat Oncol. 2018; 6: 31-38
        • Torfeh T.
        • Hammoud R.
        • Perkins G.
        • et al.
        Characterization of 3D geometric distortion of magnetic resonance imaging scanners commissioned for radiation therapy planning.
        Magn Reson Imaging. 2016; 34: 645-653
        • Price R.G.
        • Kadbi M.
        • Kim J.
        • et al.
        Technical note: Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR-SIM.
        Med Phys. 2015; 42: 5955-5960
        • Kemppainen R.
        • Vaara T.
        • Joensuu T.
        • Kiljunen T.
        Accuracy and precision of patient positioning for pelvic MR-only radiation therapy using digitally reconstructed radiographs.
        Phys Med Biol. 2018; 63: 055009
        • Korhonen J.
        • Kapanen M.
        • Sonke J.J.
        • et al.
        Feasibility of MRI-based reference images for image-guided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images.
        Acta Oncol. 2015; 54: 889-895
        • Kapanen M.
        • Collan J.
        • Beule A.
        • et al.
        Commissioning of MRI-only based treatment planning procedure for external beam radiotherapy of prostate.
        Magn Reson Med. 2013; 70: 127-135
        • Kim J.
        • Miller B.
        • Siddiqui M.S.
        • et al.
        FMEA of MR-only treatment planning in the pelvis.
        Adv Radiat Oncol. 2018; 4: 168-176
        • Korsholm M.E.
        • Waring L.W.
        • Edmund J.M.
        A criterion for the reliable use of MRI-only radiotherapy.
        Radiat Oncol. 2014; 9: 16
        • Palmer E.
        • Persson E.
        • Ambolt P.
        • et al.
        Cone beam CT for QA of synthetic CT in MRI only for prostate patients.
        J Appl Clin Med Phys. 2018; 19: 44-52
        • Korhonen J.
        • Kapanen M.
        • Keyriläinen J.
        • Seppälä T.
        • Tenhunen M.
        A dual model HU conversion from MRI intensity values within and outside of bone segment for MRI-based radiotherapy treatment planning of prostate cancer.
        Med Phys. 2014; 41: 011704
        • Dowling J.A.
        • Sun J.
        • Pichler P.
        • et al.
        Automatic substitute computed tomography generation and contouring for magnetic resonance imaging (MRI)-alone external beam radiation therapy from standard MRI sequences.
        Int J Radiat Oncol Biol Phys. 2015; 93: 1144-1153

      Comments

      Commenting Guidelines

      To submit a comment for a journal article, please use the space above and note the following:

      • We will review submitted comments as soon as possible, striving for within two business days.
      • This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
      • We require that commenters identify themselves with names and affiliations.
      • Comments must be in compliance with our Terms & Conditions.
      • Comments are not peer-reviewed.