Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer

Clinical Oncology xxx (2018) 1e10 Contents lists available at ScienceDirect

Clinical Oncology journal homepage: www.clinicaloncologyonline.net

Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer L.G.W. Kerkmeijer, M. Maspero, G.J. Meijer, J.R.N. van der Voort van Zyp, H.C.J. de Boer, C.A.T. van den Berg Department of Radiotherapy, University Medical Center Utrecht, The Netherlands Received 7 June 2018; received in revised form 29 June 2018; accepted 21 August 2018

Abstract Magnetic resonance imaging (MRI) is often combined with computed tomography (CT) in prostate radiotherapy to optimise delineation of the target and organs-at-risk (OAR) while maintaining accurate dose calculation. Such a dual-modality workflow requires two separate imaging sessions, and it has some fundamental and logistical drawbacks. Due to the availability of new MRI hardware and software solutions, CT examinations can be omitted for prostate radiotherapy simulations. All information for treatment planning, including electron density maps and bony anatomy, can nowadays be obtained with MRI. Such an MRI-only simulation workflow reduces delineation ambiguities, eases planning logistics, and improves patient comfort; however, careful validation of the complete MRI-only workflow is warranted. The first institutes are now adopting this MRI-only workflow for prostate radiotherapy. In this article, we will review technology and workflow requirements for an MRI-only prostate simulation workflow. Ó 2018 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Keywords: Dose calculation; image-guided radiotherapy; magnetic resonance imaging; MRI-only radiotherapy; position verification; prostate cancer

Current Role of MRI in External-beam Radiotherapy for Prostate Cancer In the diagnostic setting, the use of multiple functional and quantitative magnetic resonance imaging (MRI) techniques, multiparametric-MRI (mpMRI), has been advocated, leading to consensus-based guidelines for the diagnosis and reporting of prostate images on mpMRI to allow for standardisation. This has resulted in a reduction of interobserver variability and large-scale clinical implementation [1]. The Prostate Imaging and Reporting Data System (PI-RADSv2, [2]) includes clinical indications, image acquisition protocols, and a structured category assessment system. Recent studies show that MRI can be used prior to biopsy to allow for early detection of clinically significant prostate cancer and MRI-guided biopsies [3]. In addition, MRI is responsible Author for correspondence: L.G.W. Kerkmeijer, Department of Radiotherapy, University Medical Center Utrecht, Heidelberglaan 100, 3584CX Utrecht, The Netherlands. E-mail address: [email protected] (L.G.W. Kerkmeijer).

for stage migration as it is highly sensitive in the detection of extracapsular invasion or seminal vesicle infiltration [4]. This impacts treatment choice for the individual patient. Advances in imaging techniques in the last decades have impacted not only the diagnostic setting but also radiotherapy treatment strategies enormously [5]. Computed tomography (CT) is considered the primary technique in radiotherapy. Nevertheless, MRI is increasingly used in radiotherapy planning for patients with prostate cancer, especially for delineation of the target and surrounding healthy tissues, owing to its superior soft-tissue contrast compared with CT [6,7]. The differences between CT and MRI images of the prostate are illustrated in Figure 1. With CT, the boundaries of the prostate are hard to identify, whereas in MRI the prostate capsula and the internal structures are clearly visible. This is of primary importance for radiotherapy as it facilitates target delineation, which has been considered as the “major source of error in prostate external-beam radiation treatment” [9]. In addition to better visibility of the prostate and intraprostatic tumour lesions also seminal vesicles and the

https://doi.org/10.1016/j.clon.2018.08.009 0936-6555/Ó 2018 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

2

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10

Fig 1. Comparison of axial CT (left), two-dimensional (2D) T2-weighted fast spin echo MRI (centre) and 3D balanced fat-suppressed gradientecho MRI sequence (centre and right) images used in UMC Utrecht for treatment planning. In the MRI image, the internal structure of the prostate is readily apparent, while on CT bones and the intraprostatic fiducial marker are clearly visible. The figure has been reproduced from [8] with the author’s permission.

surrounding organs at risk (OAR) are more readily visible on MRI compared to CT. MRI-based prostate target contours are smaller than CT-based delineations, especially at the location of the seminal vesicles and apex of the prostate and decrease interobserver variability [9,10]. In addition, MRIbased delineations for prostate cancer may result in lower urinary toxicity due to smaller clinical target volumes (CTVs) with comparable tumour control rates [11]. As a consequence, in most clinics, the best of both techniques is combined, and MRI is used in addition to CT for the delineation of the tumour and surrounding healthy tissues [12]. MRI is primarily for contouring of the gross tumour volume (GTV), prostate and OAR (such as rectum, sphincter, bladder, penile bulb, urethra, small bowel, bony anatomy). Detection of tumour (GTV) has also opened up opportunities for focal treatment [13,14] or local dose escalation of intraprostatic lesion(s) [15e17]. This is based on the finding that local recurrences are most likely to occur at the location of an initial tumour [18]. Standardisation of GTV contouring is, however, required, and is the subject of ongoing research [19].

Rationale of MRI-only Simulation Until recently, CT has been considered imperative to acquire electron density information for dose calculations and creation of reference images to allow for X-ray or conebeam CT (CBCT)-guided radiotherapy. In most clinics, the best of both techniques is combined, and MRI is used in addition to CT for the delineation of the tumour and surrounding healthy tissues [12]. Thus, prostate cancer patients have to undergo two imaging examinationsdCT and MRIdin preparation for radiotherapy treatments. These two examinations will be taken at different time points, and consequently, the geometry of the target and OARs will vary due to varying bladder and rectal filings or small patient set-up differences. This complicates the required multimodality registration [20] and introduces ambiguities in the contouring process, most notably for OARs. This has been the rationale behind the development of a so-called MRIonly simulation [21] where all information needed for delineation, position verification, and electron density for dose calculations is derived from MRI images. Besides

avoiding delineation ambiguities, an MRI-only workflow also offers substantial advantages with respect to logistics, patient comfort, and overall costs as the CT examination can be eliminated [22,23]. The MRI-only workflow can be used both for radiotherapy planning for conventional linear accelerators (linacs; X-ray or CBCT-guided) or MRI-guided radiotherapy (MRgRT). In the case of MRgRT, not only the radiotherapy preparation phase, but also the treatment phase, are MRIguided [24]. The introduction of MRgRT makes MRI-based treatment planning increasingly important [25,26]. The present article describes the technology and adaptation required to enable an MRI-only workflow for external-beam radiotherapy for prostate cancer treatment. The focus on an MRI-only workflow is motivated by the fact that commercial solutions to generate synthetic CT images for prostate cancer are becoming available; this will facilitate clinical implementations in the near future.

The Ingredients for MRI-only Radiotherapy When MRI is used for radiotherapy, it has different requirements than for diagnostic imaging, such as geometric accuracy, imaging in the treatment position, and large field of view (FOV) coverage [27]. In addition, electron density maps for dose calculations and reference information for position verification need to be obtained from MRI images. These aspects are considered necessary ingredients for MRI-only radiotherapy and will be revised in this section. Geometric Accuracy MRI images may be compromised by geometric distortions caused by the system (system-related distortions) and the patient (patient-induced distortions) [28e30]. This may impact the accuracy of MRI-based dose calculation as well as the spatial accuracy of MRI-based delineations. To correct for system-related distortions, modern scanners are equipped with state-of-the-art gradient systems, where geometric inhomogeneities have been minimised [23]. The scanners are also equipped with software for further correction of images [27]. Displacements after corrections

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10

3

Fig 2. (a) MRI vendors are offering dedicated MRI-RT systems in which diagnostic MRI scanners are complemented with accessories, such as flat table tops with indexing, coil bridges to avoid patients body contour deformations, and positioning lasers. (b) MRI visible skin markers (e.g., Pin Point 128, Beekley Medical) can be placed at dedicated skin positions to define a patient coordinate frame. These skin positions can be marked by tattooing similar to CT simulations.

reach a maximum of 2e3 mm on the outer edges of typical volumes covered in a pelvic examination [31]. Considering that system-related distortions increase with the distance to the isocentre of the magnet of the scanner, distortions may be even lower in the proximity of a target structure, when correctly centred. Patient-induced distortions can be larger than residual system distortions after using vendorsupplied three-dimensional (3D) correction. They may be mitigated by careful choice of scan parameters, e.g., high bandwidth and low section thickness, ensuring the spatial geometric accuracy of MRI images used for delineation or synthetic-CT (sCT) generation [23,32]. In particular, it has been shown that proper choice of imaging parameters can result in dose errors caused by geometric distortions below 0.5% [33]. In any case, it is advisable to establish a quality assurance (QA) programme to monitor system-related geometric distortions [34]. Commercial QA solutions of specific MRI geometry phantoms that contain MRI visible regular grid patterns supplemented by dedicated image processing are becoming available [35]. MRI in the Treatment Position Diagnostic MRI systems are not always compatible with radiotherapy simulation [27]. Recently, MRI systems for radiotherapy purposes called “MRI-RT systems” or “MRIsimulators” [23,36]. An important adaptation has been the introduction of wide-bore (70 cm) and flat table tops that can be easily removed [37]. In addition, MRI simulators have been equipped with coil support to avoid compression of patient contours and with laser systems to facilitate the radiotherapy technicians during patient positioning [38]. Knee positioning should also be MRI compatible and the location should be reproducible on the treatment table. Analogous to CT-based radiotherapy simulation, external positioning lasers are required to reproduce the position of the patient during each treatment fraction. Instead of tattooing on the CT table, temporary skin marks in combination with MRI-visible markers can be used, which can be substituted by permanent tattoos outside and away from the magnetic field. As an alternative, as shown in Figure 2, an MRI-safe tattooing device can be used when the patient is on the table, in combination with MRI-visible skin markers.

Synthetic-CT Generation Voxel intensity of MRI images depends on proton density tissue and the nuclear magnetic relaxation [39,40]. No physical correlation has been reported between the nuclear magnetic tissue properties and the tissue electron density properties. This means that dose calculations cannot be performed directly on MRI images. Methods have been developed aiming to generate sCT images [6,41,42], which are also called “virtual CT”, “substitute CT”, or “pseudo CT”1 that should enable accurate MRI-based dose calculations. As pointed out in the first review on the topic [43], more than 50 methods have been proposed up to November 2015 when considering all the anatomical sites. Following and extending the classification of sCT generation methods proposed by Edmund & Nyholm, we classify the methods into four classes: (1) atlas-based approaches focus on aligning through registration the MRI images of a single patient to the corresponding CT images in an atlas; (2) voxel-based approaches primarily use information about voxel intensities in the MRI images to assign electron densities; (3) bulk assignment approaches, where the sCT images are generated after segmentation of MRI images in tissue classes and assignment of sCT to the classes; and (4) hybrid approaches combine categories of the voxel-, atlas-, or bulk assignment-based methods. To offer an overview of the MRI-based dose calculation evaluated up to now (May 2018) for the treatment of prostate cases, Table 1 has been compiled selecting the prostate sCT generation methods reported in the most recent reviews on the topic [43,68e70] along with the performance metrics presented by Edmund & Nyholm [43]; specifically, percent dose difference (DD), mean absolute error (MAE) and gamma analysis2 [72] were reported. For prostate

1 No agreement has been reached in literature. We adopted the “synthetic CT” in this article, considering that the acronym for pseudo-CT (pCT) can be misinterpret for planning CT, and that this nomenclature has also been adopted by the European Society for Radiation Oncology (ESTRO). 2 In case the reader may be not familiar to gamma analysis, we suggest consulting [71]. Please, also consider that the pass-rate can be calculated on different volume of interests, e.g., dose >10% or 50% of the prescription/ maximal dose and no standard is accepted. This can make difficult the comparison of several sCT methods.

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

4

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10

Table 1 Overview of sCT generation approaches for prostate radiotherapy, along with their performances in terms of percent dose difference (DD), mean absolute error (MAE), and gamma pass rate on gamma 2%, 2mm (if not differently specified). Note that dose differences are generally relative to prescribed dose or maximum dose: no general trend could be found, so we suggest the readers to consult the referenced publications for further details. This has been the rationale behind the development sCT approach

Type of radiation

Number of patients

Performance metrics

bulk bulk bulk bulk bulk bulk bulk atlas atlas atlas hybrid voxel voxel voxel voxel voxel voxel Vendors’ solutions hybrid hybrid bulkz bulkz bulkz bulkz bulkzx

g g g g g g g g g g g g g g g

5 15 10 10 39 21 10 37 39 15 39 15 10 10 15 10 35

<2 <2.5 <1 <1 1.3 <2 <1 1.5 0.3 <0.7 0.3 <1 0.4 <0.8 0.2 0.6 0.3-2

10 170 25 5 29 14 10

0.4 0.3 <0.6 0.7

36.5

0.3 <1

58 83{

DD [%]

p

g g g g g g g p

MAE [HU]

40.5 50 108 54

42 135

Reference

g2%; 2mm [%]

100 97.1 99.6/94.2* 97* 99/97* >99/93* >99 98.6/95* 93* 99.9y 99.1 100/99.1* 99.8/97.3* 98.8 98.4

[44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67]

For each approach, the classification class, type of radiation involved - photon (g) or proton (p), the number of patients included in the study, and the associated reference have been specified. The table has been reproduced from [8] with the author’s permission. * g1%;1mm . y g2%;1mm . z Bulk assignment using a model for bone segmentation. x After copying the air location from CT on the sCT images. { In the region with dose >10% of the prescription dose.

radiotherapy, sCT generation generally enables dose calculation with a deviation of maximum 2% when compared to CT-based dose calculations; however, to justify clinical use, dose deviations should be interpreted in the context of the clinically acceptable uncertainty in radiation therapy. When considering the complete radiotherapy pathway, including uncertainties in beam calibration, relative dosimetry, dose calculations, and dose delivery, the International Commission on Radiation Protection estimated uncertainty of 5% in a clinical set-up [73,74]. Dosimetric deviation of an MRIbased dose calculation (assuming CT to be the ground truth) of less than 0.5% only can be justified making up for a small fraction of the total uncertainty [62]. Recently, some sCT generation methods have been adopted in clinical practice. For example, in 2014, the Helsinki University Hospital started treating prostate cancer patients with an MRI-only radiotherapy pathway for the first time [75]. To generate sCT images, an in-house approach was developed by Korhonen and coworkers [76]. In 2016, MRI vendors and other commercial parties have

started to offer certified solutions that may enable MRIbased dose calculations3 for prostate cancer radiotherapy. The arrival of certified solutions facilitated the spread of an MRI-only workflow in a clinical setting [62,77]; however, an institution interested in introducing an MRI-only planning solution into the clinic still needs to perform an internal assessment to ensure the dosimetric accuracy of the treatment in its pathway [49,66]. During the last year, deep learning-based sCT generation was also presented relaxing the requirements for MRI image contrast (most notably bone visualisation) for sCT

3 Philips Healthcare (The Netherlands) developed a method (MR for correction attenuation, or MRCAT) announcing its European Conformity (CE) and approval from the Food and Drug Administration (FDA) on 2016-03-21 goo.gl/jtyX8H, while Spectronic Medical AB (Sweden) announced that their solution (MriPlanner) received a CE mark approval and a FDA k510 was on 2016-06-16 https://goo.gl/cY5Zrn. Siemens (Germany) offers also a certified solution (within syngo.via and called synthetic CT) from 2018-01-12 https:// goo.gl/37vEv6.

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10

generations [78e83]. Results obtained with deep learning are also of particular interest in the context of MRgRT where sCT generation should be of the order of minutes to allow daily replanning as, for example, underlined by [84]; however, so far, MRI-based dose calculations for the prostate were assessed only in the absence of a magnetic field. For MRgRT, dose calculations may require particular attention because the presence of a magnetic field affects the dose distribution in patients. This phenomenon is known as electron return effect, and it was shown that the dose distribution could be disturbed, especially when a photon beam traverses tissue interfaces [85]. So far, MRI-based dose calculation has been investigated for patients without hip implants. This limits the general adoption of the workflow as institutions need to define eligibility criteria for patient inclusion in the MRI-only pathways and be able to fall back to a CT-based pathway in case of patient ineligibility. MRI-based OAR Delineation and Protocol Optimisation In hybrid CT/MRI simulation pathways, MRI images are used to perform target delineations [7]. In an MRI-only pathway, MRI images should also be used for OAR delineations; however, the FOV acquired for diagnostic purposes is generally too small to include all the OAR. This implies that institutes should revise their MRI protocols to meet the needs of radiotherapy [27]. Revising MRI protocols can be challenging because, on the contrary to the diagnostic case [2], there is no clear consensus about which MRI sequences should be used. Moreover, given the fact that system-related geometric distortion scales with the distance to the scanner isocentre, OAR may be distorted in case attention is not dedicated to the geometric fidelity of the MRI sequences. Recently, centres involved in introducing an MRI-only pathway took the initiative of defining protocols for MRIonly-based delineations. Comparing the practices adopted by these centres (when fully reported4), it is noticeable that to enable MRI-only-based delineations, the FOV of the images should be enlarged to include contouring of all OARs as currently defined on CT images. To comply with this requirement, the FOV of T2-weighted images is extended such that delineation of target and OAR can be performed in a single sequence. This approach was suggested by an expert consensus group [86]. Moreover, other groups with interest in an MRI-only pathway followed the same approach [65,87]. Alternatively, T1-weighted gradient echo sequences have been proposed [77,88]. In general, MRI is traditionally considered as a slow imaging technique compared to, for example, CT and ultrasound [89]. A typical imaging session at an MRI system including patient set-up takes a maximum of 30e45 minutes

4 An interesting initiative regarding standardisation of MRI for use in radiotherapy that comprehends some guidelines and full report of MR protocols is within the Swedish consortium called Gentle: http:// gentleradiotherapy.se/.

5

for prostate treatment simulation [14]. An MRI-only simulation will add some time because of an extra scan (2e3 minutes) and extra positioning procedures (5e7 minutes). The duration of the scanning time should be minimised as intra-scan motion may occur [90,91] potentially undermining the advantages offered by an MRI-only pathway. Automatic contouring is not required in an MRI-only workflow but may be helpful to decrease the time required for contouring and decrease interobserver variability. Computer-aided algorithms for automatic prostate target and OAR delineation are available, which are either volume or atlas-based, and with rigid or deformable registration [92]. MRI-based Reference Information for Position Verification Set-up corrections during image-guided radiotherapy (IGRT) treatments aim to ensure that patients have identical positions between the simulation session and radiotherapy fractions [93]. An institute that introduces an MRI-only pathway needs to assess whether set-up corrections during IGRT can be performed accurately by solely utilising sCT reference images or other MRI-based surrogates. In an MRI-only pathway, MRI or sCT images are used to calculate patient set-up corrections according to the IGRT techniques adopted [94,95]. IGRT for prostate cancer is most commonly performed using intraprostatic (gold) fiducial markers (FM) as a surrogate for prostate location, and less commonly, grey values (CBCT) or bony anatomy landmarks [96e98]. Most of the MRI-only oriented investigations so far have focused on the MRI-based generation of digitally reconstructed radiographs for bone alignment [45,51,55,57,63]. It has been shown that for pelvic sites, in general, sCT can suffice as a reference image for tumour sites where position verification is based on bony anatomy registration of kilo voltage or mega voltage in-room images to a reference image [50,64,99]; however, bony anatomy is an unreliable surrogate for the prostate position [100], demanding larger planning target volume (PTV) margins compared to FM-based alignment [101e103]. Therefore, MRI-based localisation of gold FMs is crucial for accurate position verification of prostate IGRT. Markers are easily localised on CT images; however, on MRI images the appearance of markers is less distinct; markers are depicted as signal voids as they do not produce a nuclear magnetic resonance signal [104]. The appearance of FM voids varies according to imaging parameters [105] and the FM orientation along the magnetic field [106]. In an MRI-only workflow, MRI-based FM localisation could be performed manually and automatically. Few studies have investigated manual MRI-based FM localisation in any systematic way. Gustafsson et al. reported interobserver FM localisation performances can be compared with their proposed automatic method [107]. Maspero et al. performed an interobserver study comprising five radiographers showing that manual localisation can be performed with higher accuracy and precision (0.6 mm) [88]; however, this study also showed that localisation is sometimes complicated by the presence of calcifications or

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

6

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10

haemorrhage that can create signal voids similar to signal FMs [97,108]. To minimise manual interaction, automated MRI-based FM localisation methods have been proposed [107,109e111]. These methods are promising, resulting in acceptable accuracy and relatively high detection rates ranging from 84% to 96%; however, also in the case of automatic localisation, manual observation remains necessary to correct for missed detections.

Clinical Implementation of an MRI-only Workflow for Prostate Radiotherapy

Figure 3). After successful validation, the MRI-only workflow was implemented in May 2018 at the UMC Utrecht in routine clinical practice for patients with primary prostate cancer. Exclusion criteria for MRI-only radiotherapy are contraindications for MRI (as per our radiology MRI protocol) and hip prostheses. We currently use the MRI-only workflow for patients for whom FMs were implanted. At our department work is ongoing to introduce a comparable workflow for patients without FMs and to prepare for future MRI-linac treatments for prostate and other tumour localisations.

Real-Time MRgRT Before clinical introduction in the UMC Utrecht, investigations into the accuracy of MRI-only radiation planning and position verification were performed. The dosimetric accuracy of the commercially available MRI-CAT solution (Philips, Best, The Netherlands) was analysed retrospectively for 14 prostate cancer patients. Identical five-beam IMRT plans were dosimetrically evaluated on CT and sCT, respectively, demonstrating great accuracy (0.3% deviation) [66]. Furthermore, the accuracy of manual FM localisation on MRI images was evaluated by an interobserver study comprising five radiographers demonstrating high accuracy (0.6 mm) [88]. A practical hurdle has been the multi-vendor prostate treatment workflow employed at our institute for MRI, dose planning, and position verification. In particular, facilitating the forwarding of the FM positions through the workflow required software adaptation (see

The MRI-only workflow for radiotherapy simulation and planning can be used on conventional linacs, which either use megavoltage planar imaging, static kilovoltage planar imaging, CBCT or CT on rails for patient positioning on the treatment table [26,112]. To benefit most from MRI in radiotherapy, MRI should be used in an online setting (MRgRT). In MRgRT, MRI is integrated not only in the (offline) dose planning phase (MRI-only) but also as an onboard or in-room imaging for image guidance during radiotherapy treatment. The primary goal of the integration of MRI instead of X-ray or CT is to improve radiotherapy targeting accuracy, aiming to improve tumour control rates, and reducing margins striving for decreased toxicity. There are several systems commercially available or under development to facilitate MRgRT [25,26]. At the UMC

Fig 3. MRI-only workflow for prostate cancer at UMC Utrecht. First, the patient underwent MRI examinations to collect anatomical information for target and OAR delineations, generate sCT images to enable MRI-based dose calculation using MRCAT (Philips Healthcare, Finland), and locate FM. Second, as a preparatory phase to the irradiation, MRI-based target and OAR delineation are performed. As a result of the sCT images, a plan is calculated and the reference position is recorded to enable accurate patient set-up during irradiation. Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10

Utrecht, the MRI-linac has been prototyped and developed [24] in collaboration with Elekta, Philips, and other MRIlinac consortium members. Other systems are the Viewray system [113], MRI on rails [114], The Cross Cancer Institute system [115], and the Australian system [116]. Other than for off-line MRI-only radiotherapy purposes on conventional linacs, online adaptive MRgRT workflows require fast scans, adaptation, dose planning, and verification procedures. With an MRI-linac, the MRI images will be used as the positioning reference image. The clinical MRgRT workflow is currently being developed, evaluated, and implemented in several centres worldwide [26]. The rationale for MRgRT in prostate cancer has recently been published [117].

Conclusion CT and MRI are often combined in the radiotherapy simulation and planning workflow for prostate cancer. Due to innovations in MRI technology and image processing, CT information is no longer necessary for electron density maps and position-verification reference images, which can now be derived from MRI solely, allowing for an MRI-only workflow. MRI-based simulation and planning workflow requirements should be implemented, and the institute specific MRI-only workflow should be validated before clinical implementation.

Acknowledgements Research on MRI-only radiotherapy at the University Medical Center Utrecht was funded by ZonMw within the Innovative Medical Devices Initiative (IMDI) Programme, project number 1040030 (http://www.imdi.nl/project/ rasor-sharp/). M. A. Viergever and J. J.W. Lagendijk offered general support to the research. All authors declare that Elekta and Philips are members of the MRI-linac Consortium. Elekta AB financially supports all MRI-linac Consortium institutes. Philips co-founded the research project. C.A.T. van den Berg declares he is a minority shareholder of MRCode BV.

References [1] Heidenreich A, Bellmunt J, Bolla M, Joniau S, Mason M, Matveev V, et al. EAU Guidelines on Prostate Cancer. Part 1: screening, diagnosis, and treatment of clinically localised disease. Eur Urol 2011;59(1):61e71. [2] Weinreb JC, Barentsz JO, Choyke PL, Cornud F, Haider MA, Macura KJ, et al. Pi-rads prostate imaging-reporting and data system: 2015, version 2. Eur Urol 2016;69(1):16e40. [3] Hansen NL, Kesch C, Barrett T, Koo B, Radtke JP, Bonekamp D, et al. Multicentre evaluation of targeted and systematic biopsies using magnetic resonance and ultrasound imagefusion guided transperineal prostate biopsy in patients with a previous negative biopsy. Brit J Urol Int 2017;120(5): 631e638. [4] Chang AJ, Autio KA, Roach III M, Scher HI. High-risk prostate cancer classification and therapy. Nat Rev Clin Oncol 2014; 11(6):308.

7

[5] Verellen D, De Ridder M, Storme G. A (short) history of image-guided radiotherapy. Radiother Oncol 2008;86(1): 4e13. [6] Khoo VS, Dearnaley DP, Finnigan DJ, Padhani A, Tanner SF, Leach MO. Magnetic resonance imaging (MRI): considerations and applications in radiotherapy treatment planning. Radiother Oncol 1997;42(1):1e15. [7] Dirix P, Haustermans K, Vandecaveye V. The value of magnetic resonance imaging for radiotherapy planning. Semin Radiat Oncol 2014;24(3):151e159. [8] Maspero M. reportMRI-only Radiotherapy of prostate cancer. PhD thesis, Utrecht University; Utrecht. URL: http://dspace. library.uu.nl/handle/1874/362987. [9] Rasch C, Barillot I, Remeijer P, Touw A, van Herk M, Lebesque JV. Definition of the prostate in CT and MRI: a multi-observer study. Int J Radiat Oncol Biol Phys 1999;43(1): 57e66. [10] Smith WL, Lewis C, Bauman G, Rodrigues G, DSouza D, Ash R, et al. Prostate volume contouring: a 3D analysis of segmentation using 3DTRUS, CT, and MR. Int J Radiat Oncol Biol Phys 2007;67(4):1238e1247. [11] Sander L, Langkilde NC, Holmberg M, Carl J. MRI target delineation may reduce long-term toxicity after prostate radiotherapy. Acta Oncol 2014;53(6):809e814. [12] Brock KK, Dawson LA. Point: principles of magnetic resonance imaging integration in a computed tomography-based radiotherapy workflow. Semin Radiat Oncol 2014;24(3): 169e174. [13] Sciarra A, Barentsz J, Bjartell A, Eastham J, Hricak H, Panebianco V, et al. Advances in magnetic resonance imaging: how they are changing the management of prostate cancer. Eur Urol 2011;59(6):962e977. [14] Hegde JV, Mulkern RV, Panych LP, Fennessy FM, Fedorov A, Maier SE, et al. Multiparametric MRI of prostate cancer: an update on state-of-the-art techniques and their performance in detecting and localizing prostate cancer. J Magn Res Imaging 2013;37(5):1035e1054. [15] Pouliot J, Kim Y, Lessard E, Hsu IC, Vigneron DB, Kurhanewicz J. Inverse planning for HDR prostate brachytherapy used to boost dominant intraprostatic lesions defined by magnetic resonance spectroscopy imaging. Int J Radiat Oncol Biol Phys 2004;59(4):1196e1207. [16] Riches SF, Payne GS, Desouza NM, Dearnaley D, Morgan VA, Morgan SC, et al. Effect on therapeutic ratio of planning a boosted radiotherapy dose to the dominant intraprostatic tumour lesion within the prostate based on multifunctional MRI parameters. Br J Radiol 2014;87(1037):20130813. [17] Monninkhof EM, van Loon JW, van Vulpen M, Kerkmeijer LG, Pos FJ, Haustermans K, et al. Standard whole prostate gland radiotherapy with and without lesion boost in prostate cancer: toxicity in the FLAME randomized controlled trial. Radiother Oncol 2018;127(1):74e80. [18] Pucar D, Hricak H, Shukla-Dave A, Kuroiwa K, Drobnjak M, Eastham J, et al. Clinically significant prostate cancer local recurrence after radiation therapy occurs at the site of primary tumor: magnetic resonance imaging and step-section pathology evidence. Int J Radiat Oncol Biol Phys 2007;69(1): 62e69. [19] van Schie MA, Dinh CV, van Houdt PJ, Pos FJ, Heijmink SW, Kerkmeijer LG, et al. Contouring of prostate tumors on multiparametric MRI: evaluation of clinical delineations in a multicenter radiotherapy trial. Radiother Oncol 2018;128(2): 321e326. [20] Nyholm T, Nyberg M, Karlsson MG, Karlsson M. Systematisation of spatial uncertainties for comparison between a

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

8

[21]

[22]

[23] [24]

[25]

[26]

[27] [28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10 MRI and a CT-based radiotherapy workflow for prostate treatments. Radiat Oncol 2009;4(1):1e9. Fraass BA, McShan DL, Diaz RF, Ten Haken RK, Aisen A, Gebarski S, et al. Integration of magnetic resonance imaging into radiation therapy treatment planning: i. technical considerations. Int J Radiat Oncol Biol Phys 1987;13(12): 1897e1908. Karlsson M, Karlsson MG, Nyholm T, Amies C, Zackrisson B. Dedicated magnetic resonance imaging in the radiotherapy clinic. Int J Radiat Oncol Biol Phys 2009;74(2):644e651. Devic S. MRI simulation for radiotherapy treatment planning. Med Phys 2012;39(11):6701. Raaymakers BW, Raaijmakers AJE, Kotte ANTJ, Jette D, Lagendijk JJW. Integrating a MRI scanner with a 6 MV radiotherapy accelerator: dose deposition in a transverse magnetic field. Phys Med Biol 2004;49(17):4109e4118. nard C, van der Heide UA. Introduction: magnetic resoMe nance imaging comes of age in radiation oncology. Semin Radiat Oncol 2014;24(3):149e150. Pollard JM, Wen Z, Sadagopan R, Wang J, Ibbott GS. The future of image-guided radiotherapy will be MRI guided. Br J Radiol 2017;90(1073):20160667. Schmidt MA, Payne GS. Radiotherapy planning using MRI. Phys Med Biol 2015;60(22):R323eR361. Jezzard P. Physical basis of spatial distortions in magnetic resonance images. In: Bankman IN, editor. Handbook of medical imaging. Orlando, FL: Academic Press; 2000. p. 425e438. €tter R. Aspects of MRI image disFransson A, Andreo P, Po tortions in radiotherapy treatment planning. Strahlenther Onkol 2001;177(2):59e73. Walker A, Liney G, Metcalfe P, Holloway L. MRI distortion: considerations for MRI based radiotherapy treatment planning. Aus Phys Eng Sci Med 2014;37(1):103e113. Wang D, Strugnell W, Cowin G, Doddrell DM, Slaughter R. Geometric distortion in clinical MRI systems. Part I: evaluation using a 3D phantom. Mag Res Imaging 2004;22(9): 1211e1221. Stanescu T, Wachowicz K, Jaffray DA. Characterization of tissue magnetic susceptibility-induced distortions for MRIgRT. Med Phys 2012;39(12):7185e7193. Adjeiwaah M, Bylund M, Lundman JA, Thellenberg Karlsson C, Jonsson JH, Nyholm T. Quantifying the effect of 3T MRI residual system and patient-induced susceptibility distortions on radiotherapy treatment planning for prostate cancer. Int J Radiat Oncol Biol Phys 2017;100(2):317e324. Paulson ES, Erickson B, Schultz C, Allen Li X. Comprehensive MRI simulation methodology using a dedicated MRI scanner in radiation oncology for external-beam radiation treatment planning. Med Phys 2015;42(1):28e39. Kapanen M, Collan J, Beule A, Seppl T, Saarilahti K, Tenhunen M. Commissioning of MRI-only based treatment planning procedure for external-beam radiotherapy of prostate. Magn Reson Med 2013;70(1):127e135. Metcalfe P, Liney GP, Holloway L, Walker A, Barton M, Delaney GP, et al. The potential for an enhanced role for MRI in radiation-therapy treatment planning. Technol Cancer Res Treat 2013;12(5):429e446. Mcjury M, O’Neill A, Lawson M, Mcgrath C, Grey A, Page W, et al. Assessing the image quality of pelvic MRI images acquired with a flat couch for radiotherapy treatment planning. Br J Radiol 2011;84(1004):750e755. Low DA. MRI guided radiotherapy. Cancer treatment and research. In: Wong JYC, Radany EH, Schultheiss TE, editors.

[39]

[40]

[41]

[42]

[43] [44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

Advances in radiation Oncology. New York: Springer; 2017. p. 41e67. McRobbie DW, Moore EA, Graves MJ, Prince MR. MRI from picture to proton. Cambridge: Cambridge University Press; 2006. ISBN 9780511545405. Brown RW, Cheng YCN, Haacke EM, Thompson MR, Venkatesan R. Magnetic resonance imaging: physical properties and sequence design. Wiley; 2014. ISBN 0471720852. Beavis AW, Gibbs P, Dealey RA, Whitton VJ. Radiotherapy treatment planning of brain tumours using MRI alone. Br J Radiol 1998;71(845):544e548. Ramsey CR, Oliver AL. Magnetic resonance imaging based digitally reconstructed radiographs, virtual simulation, and three-dimensional treatment planning for brain neoplasms. Med Phys 1998;25(10):1928e1934. Edmund JM, Nyholm T. A review of substitute CT generation for MRI-only radiation therapy. Radiat Oncol 2017;12(1):28. Lee YK, Bollet M, Charles-Edwards G, Flower MA, Leach MO, McNair H, et al. Radiotherapy treatment planning of prostate cancer using magnetic resonance imaging alone. Radiother Oncol 2003;66(2):203e216. Chen L, Price RA, Nguyen TB, Wang L, Li JS, Qin L, et al. Dosimetric evaluation of MRI-based treatment planning for prostate cancer. Phys Med Biol 2004;49(22):5157e5170. Eilertsen K, Nilsen Tor Arne Vestad L, Geier O, Skretting A. A simulation of MRI based dose calculations on the basis of radiotherapy planning CT images. Acta Oncol 2008;47(7): 1294e1302. Jonsson JH, Karlsson MG, Karlsson M, Nyholm T. Treatment planning using MRI data: an analysis of the dose calculation accuracy for different treatment regions. Radiat Oncol 2010; 5(1):62. Lambert J, Greer PB, Menk F, Patterson J, Parker J, Dahl K, et al. MRI-guided prostate radiation therapy planning: Investigation of dosimetric accuracy of MRI-based dose planning. Radiother Oncol 2011;98(3):330e334. Korsholm ME, Waring LW, Edmund JM. A criterion for the reliable use of MRI-only radiotherapy. Radiat Oncol 2014; 9(16). Doemer A, Chetty IJ, Glide-Hurst C, Nurushev T, Hearshen D, Pantelic M, et al. Evaluating organ delineation, dose calculation and daily localization in an open-MRI simulation workflow for prostate cancer patients. Radiat Oncol 2015; 10(1):37. Dowling JA, Lambert J, Parker J, Salvado O, Fripp J, Capp A, et al. An atlas-based electron density mapping method for magnetic resonance imaging (MRI)-alone treatment planning and adaptive MRI-based prostate radiation therapy. Int J Radiat Oncol Biol Phys 2012;83(1):e5ee11. nault D, Ghose S, Dowling JA, Sun J, Pichler P, Rivest-He Richardson H, 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(5):1144e1153. Guerreiro F, Burgos N, Dunlop A, Wong K, Petkar I, Nutting C, et al. Evaluation of a multi-atlas CT synthesis approach for MRI-only radiotherapy treatment planning. Phys Med 2017;35:7e17. Ghose S, Greer PB, Sun J, Pichler P, Rivest-Henault D, Mitra J, et al. Regression and statistical shape model based substitute CT generation for MRI alone external-beam radiation therapy from standard clinical MRI sequences. Phys Med Biol 2017;62(22):8566e8580.

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10 [55] Kim J, Glide-Hurst C, Doemer A, Wen N, Movsas B, Chetty IJ. Implementation of a novel algorithm for generating synthetic CT images from magnetic resonance imaging data sets for prostate cancer radiation therapy. Int J Radiat Oncol Biol Phys 2015;91(1):39e47. [56] Andreasen D, Van Leemput K, Edmund JM. A patch-based pseudo-CT approach for MRI-only radiotherapy in the pelvis. Med Phys 2016;43(8Part1):4742e4752. €inen J, Seppa € l€ [57] Korhonen J, Kapanen M, Keyrila a T, Tenhunen M. A dual model HU conversion from MRI intensity values within and outside of bone segment for MRIbased radiotherapy treatment planning of prostate cancer. Med Phys 2014;41(1). 011704. €inen J, Seppa €la € T, [58] Korhonen J, Kapanen M, Keyrila Tuomikoski L, Tenhunen M. Influence of MRI-based bone outline definition errors on external radiotherapy dose calculation accuracy in heterogeneous pseudo-CT images of prostate cancer patients. Acta Oncol 2014b;53(8):1100e1106. [59] Koivula L, Wee L, Korhonen J. Feasibility of MRI-only treatment planning for proton therapy in brain and prostate cancers: dose calculation accuracy in substitute CT images. Med Phys 2016;43(8Part1):4634e4642. €la € T, Collan J, Dowling JA, [60] Koivula L, Kapanen M, Seppa Greer PB, et al. Intensity-based dual model method for generation of synthetic CT images from standard T2weighted MRI imagesdgeneralized technique for four different MRI scanners. Radiother Oncol 2017;125(3): 411e419. € m F, Nilsson T, Nyholm T, Jonsson J, [61] Siversson C, Nordstro Gunnlaugsson A, et al. Technical Note: MRI only prostate radiotherapy planning using the statistical decomposition algorithm. Med Phys 2015;42(10):6090e6097. €m F, Sohlin M, [62] Persson E, Gustafsson C, Nordstro Gunnlaugsson A, Petruson K, et al. MRI-OPERAeA multicenter/multivendor validation of MRI-only prostate treatment planning using synthetic CT images. Int J Radiat Oncol Biol Phys 2017;99(3):692e700. [63] Tyagi N, Fontenla S, Zhang J, Cloutier M, Kadbi M, Mechalakos J, et al. Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis. Phys Med Biol 2016;62(8):2961e2975. [64] Kemppainen R, Suilamo S, Tuokkola T, Lindholm P, €inen J. Magnetic resonance-only simulaDeppe MH, Keyrila tion and dose calculation in external-beam radiation therapy: a feasibility study for pelvic cancers. Acta Oncol 2017; 56(6):792e798. [65] Christiansen RL, Jensen HR, Brink C. Magnetic resonance only workflow and validation of dose calculations for radiotherapy of prostate cancer. Acta Oncol 2017;56(6): 787e791. [66] Maspero M, Seevinck PR, Schubert G, Hoesl MAU, van Asselen B, Viergever MA, et al. Quantification of confounding factors in MRI-based dose calculations as applied to prostate IMRT. Phys Med Biol 2017;62(3):948e965. [67] Maspero M, van den Berg CAT, Landry G, Belka C, Parodi K, Seevinck PR, et al. Feasibility of MRI-only proton dose calculations for prostate cancer radiotherapy using a commercial pseudo-CT generation method. Phys Med Biol 2017; 62(24):9159e9176. [68] Johnstone E, Wyatt JJ, Henry AM, Short SC, SebagMontefiore D, Murray L, et al. A systematic review of synthetic CT generation methodologies for use in MRI-only radiotherapy. Int J Radiat Oncol Biol Phys 2017;100(1): 199e217.

9

[69] Owrangi AM, Greer PB, Glide-Hurst CK. MRI-only treatment planning: benefits and challenges. Phys Med Biol 2018;63(5). 05TR01. [70] Wafa B, Moussaoui A. A review on methods to estimate a ct from MRI data in the context of MRI-alone RT. Med Tech J 2018;2(1):150e178. [71] Low DA. Gamma dose distribution evaluation tool. J Phys 2010;250(1). 012071. [72] Low DA, Harms WB, Mutic S, Purdy JA. A technique for the quantitative evaluation of dose distributions. Med Phys 1998; 25(5):656e661. [73] ICRP. Prevention of accidents to patients undergoing radiation therapy. ICRP Publication 86. Ann ICRP 2000;30(3). [74] Thwaites D. Accuracy required and achievable in radiotherapy dosimetry: have modern technology and techniques changed our views? J Phys 2013;444(1). 012006. €la € T, Koivula L, [75] Tenhunen M, Korhonen J, Kapanen M, Seppa Collan J, et al. MRI-only based radiation therapy of prostate cancer: workflow and early clinical experience. Acta Oncol 2018;0(0):1e6. € H, Sepp€ € T, Kapanen M, Saarilahti K, [76] Korhonen J, Visap€ aa ala Tenhunen M. Clinical experiences of treating prostate cancer patients with magnetic resonance imaging-only based radiation therapy treatment planning workflow. Int J Radiat Oncol Biol Phys 2016;96(2):S225. [77] Tyagi N, Fontenla S, Zelefsky M, Chong-Ton M, Ostergren K, Shah N, et al. Clinical workflow for MRI-only simulation and planning in prostate. Radiat Oncol 2017;12(1):119. [78] Nie D, Trullo R, Petitjean C, Ruan S, Shen D. Medical image synthesis with context-aware generative adversarial networks. IEEE 2016. arXiv:1612.05362. [79] Han X. MRI-based synthetic CT generation using a deep convolutional neural network method. Med Phys 2017;44(4): 1408e1419. [80] Wolterink JM, Dinkla AM, Savenije MHF, Seevinck PR, van den Berg CAT, Isgum I. Deep MRI to CT synthesis using unpaired data. In: Tsaftaris SA, Gooya A, Frangi AF, Prince JL, editors. Simulation and synthesis in medical imaging. Cham: Springer International Publishing; 2017. p. 14e23. [81] Savenije MHF, Maspero M, Dinkla MA, Seevinck PR, van den Berg CAT. E37-1159: MRI-based synthetic CT with conditional Generative Adversarial Network for prostate RT planning. In: ESTRO 37 2018. p. 5150e5151. [82] Maspero M, Savenije MHF, Dinkla AM, Seevinck PR, Intven MPW, Juergenliemk-Schulz IM, et al. Dose evaluation of fast synthetic-CT generation using a generative adversarial network for general pelvis MR-only radiotherapy. Phys Med Biol 2018 Aug 15;63(18). 185001-13. PubMed PMID: [30109989], http://iopscience.iop.org/article/10.1088/13616560/aada6d. [83] Dinkla AM, Wolterink JM, Maspero M, Savenije MH, Verhoeff JJ, Seravalli E, et al. MRI-only brain radiotherapy: dosimetric evaluation of synthetic CTs generated by a dilated convolutional neural network. Int J Radiat Oncol Biol Phys 2018;0(0). in press. € rgenliemk-Schulz IM, Bol GH, Glitzner M, [84] Raaymakers BW, Ju Kotte ANTJ, van Asselen B, et al. First patients treated with a 1.5 T MRI-Linac: clinical proof of concept of a high-precision, high-field MRI guided radiotherapy treatment. Phys Med Biol 2017;62(23):L41eL50. [85] Raaijmakers AJE, Raaymakers BW, Lagendijk JJW. Magneticfield-induced dose effects in MRI-guided radiotherapy systems: dependence on the magnetic field strength. Phys Med Biol 2008;53(4):909e923.

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009

10

L.G.W. Kerkmeijer et al. / Clinical Oncology xxx (2018) 1e10

[86] Paulson ES, Crijns SPM, Keller BM, Wang J, Schmidt MA, Coutts G, et al. Consensus opinion on MRI simulation for external-beam radiation treatment planning. Radiother Oncol 2016;121(2):187e192. [87] Tyagi N, Fontenla S, Zhang J, Cloutier M, Kadbi M, Mechalakos J, et al. Dosimetric and workflow evaluation of first commercial synthetic CT software for clinical use in pelvis. Phys Med Biol 2017;62(8):2961e2975. [88] Maspero M, Seevinck PR, Willems NJ, Sikkes GG, de Kogel GJ, de Boer HCJ, et al. Evaluation of manual gold fiducial marker localisation for magnetic resonance-only prostate radiotherapy. Radiat Oncol 2018;13(105). [89] Brix L, Sørensen TS, Berber Y, Ries M, Stausbøl-Grøn B, Ringgaard S. Feasibility of interactive magnetic resonance imaging of moving anatomy for clinical practice. Clin Physiol Funct Imaging 2014;34(1):32e38. [90] McPartlin AJ, Li X, Kershaw LE, Heide U, Kerkmeijer L, Lawton C, et al. MRI-guided prostate adaptive radiotherapy d a systematic review. Radiother Oncol 2016;119(3): 371e380. [91] Tree A, Khoo V, van As N. Prostate motion: Implications and management strategies. In: Robotic radiosurgery. Treating prostate cancer and related genitourinary applications. Cham: Springer; 2012. p. 51e65. [92] Alvarez C, Mart’ınez F, Romero E. A multiresolution prostate representation for automatic segmentation in magnetic resonance images. Med Phys 2017;44(4):1312e1323. [93] Verellen D, Ridder MD, Linthout N, Tournel K, Soete G, Storme G. Innovations in image-guided radiotherapy. Nat Rev Cancer 2007;7(12):949e960. [94] Das S, Liu T, Jani AB, Rossi P, Shelton J, Shi Z, et al. Comparison of image-guided radiotherapy technologies for prostate cancer. Am J Clin Oncol 2014;37:616e623. [95] Nabavizadeh N, Elliott DA, Chen Y, Kusano AS, Mitin T, Thomas CR, et al. Image guided radiation therapy (IGRT) practice patterns and IGRT’s impact on workflow and treatment planning: results from a National Survey of American Society for Radiation Oncology Members. Int J Radiat Oncol Biol Phys 2016;94(4):850e857. [96] Bujold A, Craig T, Jaffray D, Dawson LA. Image-guided radiotherapy: has it influenced patient outcomes? Semin Radiat Oncol 2012;22(1):50e61. [97] Ng M, Brown E, Williams A, Chao M, Lawrentschuk N, Chee R. Fiducial markers and spacers in prostate radiotherapy: current applications. BJU Int 2014;113(S2):13e20. [98] Zaorsky NG, Showalter TN, Ezzell GA, Nguyen PL, Assimos DG, D’Amico AV, et al. ACR appropriateness criteria R external-beam radiation therapy treatment planning for clinically localized prostate cancer, part I of II. Adv Radiat Oncol 2017;2(1):62e84. €inen J, [99] Korhonen J, Kapanen M, Sonke JJ, Wee L, Salli E, Keyrila et al. Feasibility of MRI-based reference images for imageguided radiotherapy of the pelvis with either cone-beam computed tomography or planar localization images. Acta Oncol 2015;54(6):889e895. [100] Schallenkamp JM, Herman MG, Kruse JJ, Pisansky TM. Prostate position relative to pelvic bony anatomy based on intraprostatic gold markers and electronic portal imaging. Int J Radiat Oncol Biol Phys 2005;63(3):800e811. [101] Beard CJ, Kijewski P, Bussi`ere M, Gelman R, Gladstone D, Shaffer K, et al. Analysis of prostate and seminal vesicle motion: implications for treatment planning. Int J Radiat Oncol Biol Phys 1996;34(2):451e458.

[102] van Herk M, Bruce A, Kroes AP, Shouman T, Touw A, Lebesque. Quantification of organ motion during conformal radiotherapy of the prostate by three dimensional image registration. Int J Radiat Oncol Biol Phys 1995;33(5): 1311e1320. [103] Greer P, Dahl K, Ebert M, Wratten C, White M, Denham J. Comparison of prostate set-up accuracy and margins with off-line bony anatomy corrections and online implanted fiducial-based corrections. J Med Imaging Radiat Oncol 2008; 52(5):511e516. [104] Zangger K, Armitage LM. Silver and gold NMR. Metal-based Drugs 1999;6(4e5):239e245. [105] Lim TY, Kudchadker RJ, Wang J, Stafford RJ, MacLellan C, Rao A, et al. Effect of pulse sequence parameter selection on signal strength in positive-contrast MRI markers for MRIbased prostate postimplant assessment. Med Phys 2016; 43(7):4312e4322. [106] Jonsson JH, Garpebring A, Karlsson MG, Nyholm T. Internal fiducial markers and susceptibility effects in MRI: simulation and measurement of spatial accuracy. Int J Radiat Oncol Biol Phys 2012;82(5):1612e1618. [107] Gustafsson C, Korhonen J, Persson E, Gunnlaugsson A, Nyholm T, Olsson LE. Registration free automatic identification of gold fiducial markers in MRI target delineation images for prostate radiotherapy. Med Phys 2017;44(11): 5563e5574. [108] Hong CG, Yoon BI, Choe HS, Ha US, Sohn DW, Cho YH. The prevalence and characteristic differences in prostatic calcification between health promotion center and urology department outpatients. Kor J Urol 2012;53(5):330. nault D, Fazlollahi A, Stanwell P, [109] Ghose S, Mitra J, Rivest-He Pichler P, et al. MRI-alone radiation therapy planning for prostate cancer: automatic fiducial marker detection. Med Phys 2016;43(5):2218e2228. [110] Dinis Fernandes C, Dinh CV, Steggerda MJ, ter Beek LC, Smolic M, van Buuren LD, et al. Prostate fiducial marker detection with the use of multi-parametric magnetic resonance imaging. Phys Imag Radiat Oncol 2017;1:14e20. [111] Maspero M, van den Berg CAT, Zijlstra F, Sikkes GG, de Boer HCJ, Meijer GJ, et al. Evaluation of an automatic MRbased gold fiducial marker localisation method for MR-only prostate radiotherapy. Phys Med Biol 2017;62(20): 7981e8002. [112] Kupelian P, Meyer JL. Image-guided, adaptive radiotherapy of prostate cancer: toward new standards of radiotherapy practice. Front Radiat Ther Oncol 2011;43:344e368. [113] Mutic S, Dempsey JF. The ViewRay system: magnetic resonance-guided and controlled radiotherapy. Sem Radiat Oncol 2014;24(3):196e199. [114] Jaffray DA, Carlone MC, Milosevic MF, Breen SL, Stanescu T, Rink A, et al. A facility for magnetic resonance-guided radiation therapy. Semin Radiat Oncol 2014;24(3):193e195. [115] Fallone BG. The rotating biplanar linac-magnetic resonance imaging system. Semin Radiat Oncol 2014;24(3):200e202. [116] Keall PJ, Barton M, Crozier S. The Australian magnetic resonance imaging-linac program. Semin Radiat Oncol 2014;24(3): 203e206. [117] Pathmanathan AU, van As NJ, Kerkmeijer LGW, Christodouleas J, Lawton CAF, Vesprini D, et al. Magnetic resonance imaging-guided adaptive radiation therapy: a “game changer” for prostate treatment? Int J Radiat Oncol Biol Phys 2018;100(2):361e373.

Please cite this article in press as: Kerkmeijer LGW, et al., Magnetic Resonance Imaging only Workflow for Radiotherapy Simulation and Planning in Prostate Cancer, Clinical Oncology (2018), https://doi.org/10.1016/j.clon.2018.08.009