A multi-institutional dosimetry audit of rotational intensity-modulated radiotherapy

Radiotherapy and Oncology 113 (2014) 272–278

Contents lists available at ScienceDirect

Radiotherapy and Oncology journal homepage: www.thegreenjournal.com

Quality assurance

A multi-institutional dosimetry audit of rotational intensity-modulated radiotherapy Catharine H. Clark a,b,c,⇑, Mohammad Hussein a,d, Yatman Tsang c,e, Russell Thomas b, Dean Wilkinson c,e, Graham Bass b, Julia Snaith b, Clare Gouldstone b, Steve Bolton f,g, Rebecca Nutbrown b, Karen Venables c,e, Andrew Nisbet a,d a Department of Medical Physics, Royal Surrey County Hospital, Guildford; b National Physical Laboratory, London; c NCRI Radiotherapy Trials Quality Assurance Group; d Centre for Nuclear and Radiation Physics, University of Surrey; e Mount Vernon Hospital, Northwood, London; f Christie Hospital, Manchester; and g Institute of Physics and Engineering in Medicine, York, UK

a r t i c l e

i n f o

Article history: Received 12 May 2014 Received in revised form 4 November 2014 Accepted 6 November 2014 Available online 23 November 2014 Keywords: Audit IMRT VMAT Detector arrays

a b s t r a c t Background: Rotational IMRT (VMAT and Tomotherapy) has now been implemented in many radiotherapy centres. An audit to verify treatment planning system modelling and treatment delivery has been undertaken to ensure accurate clinical implementation. Material and methods: 34 institutions with 43 treatment delivery systems took part in the audit. A virtual phantom planning exercise (3DTPS test) and a clinical trial planning exercise were planned and independently measured in each institution using a phantom and array combination. Point dose differences and global gamma index (c) were calculated in regions corresponding to PTVs and OARs. Results: Point dose differences gave a mean (±sd) of 0.1 ± 2.6% and 0.2 ± 2.0% for the 3DTPS test and clinical trial plans, respectively. 34/43 planning and delivery combinations achieved all measured planes with >95% pixels passing c < 1 at 3%/3 mm and rose to 42/43 for clinical trial plans. A statistically significant difference in c pass rates (p < 0.01) was seen between planning systems where rotational IMRT modelling had been designed for the manufacturer’s own treatment delivery system and those designed independently of rotational IMRT delivery. Conclusions: A dosimetry audit of rotational radiotherapy has shown that TPS modelling and delivery for rotational IMRT can achieve high accuracy of plan delivery. Ó 2014 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 113 (2014) 272–278

Intensity Modulated Radiotherapy (IMRT), and more recently, Volumetric Modulated Arc Therapy (VMAT) and Tomotherapy are now implemented in many radiotherapy clinics [1]. Patient specific quality assurance for these rotational IMRT (RIMRT) techniques can be carried out using a second calculation with independent software or making a dose measurement on the linac, which has the added benefit of verifying the leaf motions, dose rates and, if applicable, gantry motion. ESTRO guidelines on the verification of IMRT have previously recommended that ‘‘more information is urgently needed about the accuracy of IMRT treatment delivery by having similar types of independent audit or inter-comparison programmes’’ as those published [2–5]. Since the commercial introduction of rotational IMRT capability on conventional linear accelerators, there has been a rapid uptake of VMAT and

⇑ Corresponding author at: Department of Medical Physics, St Luke’s Cancer Centre, Royal Surrey County Hospital, Egerton Road, Guildford, Surrey GU2 7XX, UK. E-mail address: [email protected] (C.H. Clark). http://dx.doi.org/10.1016/j.radonc.2014.11.015 0167-8140/Ó 2014 Elsevier Ireland Ltd. All rights reserved.

Tomotherapy, such that in 2010 in the United Kingdom (UK) around 30% of centres were already treating with some form of RIMRT. This led to the need for an audit of RIMRT with the following aims: independent verification of the implementation, investigation of the capability of the planning and delivery systems, assessment of whether each planning and delivery system had been optimised uniformly across each institution and credentialing for use of RIMRT in clinical trials. This external dosimetry audit focussed entirely on rotational IMRT. Two previous studies included a few VMAT and Tomotherapy systems in IMRT audits [6,7] whereas others have been entirely static gantry IMRT [3,5,8,9]. This audit made use of a commercial detector array which allows the verification of dose in a large number of positions with immediate results. Our previous study, developed a methodology for using such a system in radiotherapy audits of rotational IMRT by comparing against other conventional systems of dosimetry such as film, ion chambers and alanine [10].

C.H. Clark et al. / Radiotherapy and Oncology 113 (2014) 272–278

Materials and methods All institutions in the UK who were already treating patients with a rotational IMRT technique or were ready to start treating by March 2013 were included in this audit. These comprised of 25 institutions with Varian linacs (Varian Medical Systems, Inc., Palo Alto, CA), 12 with Elekta linacs (Elekta AB, Stockholm, Sweden) and 6 with Helical Tomotherapy systems (Accuray-TomoTherapy, Madison, WI). The measurement system was the PTW Octavius II and seven29 array (PTW-Freiburg GmbH, Germany) and was chosen as it was robust, relatively easy to handle for transportation, straight forward to calibrate and gave analysis results typical of systems used in visited hospitals [11]. It has also been compared to different systems to prove its ability to detect errors [12]. This equipment was transported to each institution. Plans Each institution was asked to create two treatment plans. The first was a generic plan which had been designed for the purpose of comparing all rotational IMRT techniques, called the 3DTPS test [13], which is a virtual phantom with pre-delineated volumes (see Supplementary Fig. 1). The test included five PTVs and one OAR, each of which has different specified dose constraints per fraction (2.5 Gy: primary PTV2, 2.0 Gy: PTV3 and PTV5, and 1.5 Gy: PTV1 and PTV4, with maximum dose to the OAR less than 1.0 Gy). The plan was validated on each of the planning systems before the audit began and was designed to be challenging, and, as much as possible, equally so on each system, in terms of ability to achieve mandatory and optional dose constraints [13]. The second plan was chosen from amongst three different clinical sites of prostate and pelvic nodes (PPN), head and neck or breast from the pre-trial planning exercises of the national clinical trials portfolio (NCRI). The clinical plan was created on pre-delineated CT datasets using the local planning protocol. This also gave the institution the opportunity to fulfil part of the credentialing programme requirements to join the specific trial [14]. Both the 3DTPS plan and selected clinical plan had to be submitted and reviewed by the audit team before the visit could take place. Verification plan creation Each institution was provided with a set of CT scans of the Octavius II phantom. They were also provided with CT number to relative electron density and mass density calibration curves and were instructed to import the curve into their TPS where appropriate. This was not a mandatory step as the uncertainty was estimated to be within 0.5% [10]. Each institution was instructed to apply their normal procedure for couch correction; e.g. inserting or ignoring a couch structure in the planning system. For the 3DTPS plan, institutions were given detailed instructions to ensure that the position of the dose distribution relative to the phantom was consistently reproduced, and thus the dose planes were measured in the same part of the plan from institution-toinstitution. These were two coronal (horizontal) planes and a sagittal (vertical) plane (see Supplementary Fig. 1). The first coronal plane directly intersected PTV1, PTV2, PTV4 and PTV5. The second coronal plane was 4 cm posterior with respect to the first, and intersected the OAR, PTV1 and PTV3. A couch vertical shift was employed to transfer between the two setups and the same displacement for dose prediction calculations was made using the TPS. The sagittal plane intersected PTV2, PTV4, PTV5 and the OAR. A re-orientation of the phantom was employed for this third setup. For the clinical trial plans, institutions were given similar instructions as to how to transfer and position the plan on the scan of the verification phantom. One coronal and one sagittal plane

273

were measured, to sample the main PTVs and OARs (see Supplementary Fig. 2). All verification plans were submitted for independent evaluation using the Visualization and Organization of Data for Cancer Analysis (VODCA) independent evaluation software version 4.3.0 [15]. For all plans, DICOM dose cubes were exported to be used for analysis.

Analysis Analysis of each measured dose plane was made using the PTW Verisoft software (v 5.0). Absolute global gamma (c) index calculations [16] were made which combine distance to agreement with a dose difference for every pixel in the plane against the 3D TPS dose distribution using the point dose spacing equal to that of the array (1 cm). A dose point, chosen in a high dose, low gradient region, was nominally set as 100% and a 20% threshold was applied to remove the low dose peripheral region. A range of gamma parameters were calculated, from 2 to 4% dose difference and from 2 to 4 mm distance to agreement. Differences between dose points measured in individual array cells and TPS predicted points were calculated, in regions relating to PTVs and OARs, as (Dmeas DTPS/DTPS). For the 3DTPS test six separate point dose locations were chosen, to sample different dose levels in the 3DTPS test. These were a central point in PTV2 and a point within PTV1 in the first coronal plane, and points in the PTV2 and PTV3 in the second coronal plane. In the sagittal orientation, a point was recorded in the PTV2 and in the OAR. For the clinical plans: Three to four separate point dose locations were chosen, to sample different dose levels in the PTVs and OARs, with location dependent on the clinical trial plan. The TPS were grouped according to whether RIMRT modelling had been specifically designed for the manufacturer’s own treatment delivery system (Type 1: Eclipse and HiArt) or had been designed to be independent of vendor or RIMRT delivery (Type 2: Monaco, OMP and Pinnacle). The data was also analysed by delivery system (Varian, Elekta and Tomotherapy).

Results Measurements were made in 34 institutions in the UK with 43 different planning and delivery combinations. In total 215 dose planes were measured, and 413 point dose differences were calculated. Fig. 1 shows the spread of the point dose differences in all points measured in the 3DTPS and clinical plans. The outliers were mainly measured in regions corresponding to OARs. For outliers corresponding to PTVs, the other PTVs measured in the plan were generally within 1sd of the mean for both the 3DTPS and the clinical plans. For the gamma index calculations, Table 1 shows the mean pass rate for a range of c parameters calculated per measured plane for the 3DTPS plan and for the clinical plans by site. The percentage of planes achieving at least 95% of c < 1 are also shown. For the 3DTPS plan, 34/43(79.1%) of planning and delivery combinations achieved all measured planes with >95% pixels passing c < 1 at 3%/3 mm with 12/43(27.9%) planning and delivery combinations passing all three measured planes with 100%, see Fig. 2. Combination 31 had only commissioned their system for simple prostate plans and therefore the 3DTPS was significantly more challenging than their routine plans (shown by the coronal plane pass rates of 85.5%(3DTPS) and 98.6%(prostate)). Repeat measurements were made in centres 29, 30, 31 and 41 (final results shown). In combination 25 and 41 two of the planes were >95%, and in 8 and 35 one plane was >95%. Apart from combination 29, all these results were from Monaco, Pinnacle and OMP TPS (Type 2). 27(62.7%) and

274

Rotational IMRT dosimetry audit

Fig. 1. The point dose differences between the measured and TPS predictions for the 3DTPS plan for (a) all dose points (b) points in PTVs only; and for the clinical plans (c) all dose points and (d) points in PTVs only.

Table 1 Mean gamma pass rate and percentage of planes achieving at least 95% of c < 1 for the 3DTPS test and clinical plans by site. Two planes (a sagittal and a coronal) were measured for each plan. Site (no. plans)

Breast (1) Prostate and pelvic nodes (15) Head and neck (27) 3DTPS (43)

Mean c pass rate (%)

Percentage of planes achieving at least 95% of c < 1

2%/2 mm

3%/2 mm

3%/3 mm

2%/2 mm

3%/2 mm

3%/3 mm

99.8 94.9 93.4 91.5

100 98.4 97.8 96.3

100 99.5 99.2 98.3

100 73.1 55.4 56.1

100 88.5 85.7 75.0

100 100 98.2 88.6

17(39.5%) combinations achieved all measured planes with >95% pixels passing c < 1 at 3%/2 mm and 2%/2 mm, respectively. For the clinical trial plans, 42(97.7%), 30(69.8%) and 23(53.5%) out of the 43 planning and delivery combinations passed all planes with more than 95% of pixels passing, at 3%/3 mm, 3%/2 mm and 2%/2 mm respectively. The combination which did not pass at 3%/3 mm also did not achieve all three planes of the 3DTPS with more than 95% of pixels passing these gamma criteria. Overall pass rates were seen to be highest for breast then PPN, then head and neck, although only one breast plan was measured (see Table 1). Gamma pass rates at 3%/3 mm, 3%/2 mm and 2%2 mm are shown for each individual planning system and delivery system in Table 2. For all TPS and delivery systems 100% pass rate could be achieved for 3%/3 mm for both the 3DTPS and clinical plans. For 3%/2 mm, 100% pass rate could be achieved in 3/5 TPS and in all delivery systems for all plans. Fig. 3 shows boxplots of the pass

rates for a range of gamma criteria. A statistically significant difference of p < 0.01 (using a Mann–Whitney test) was seen between Type 1 and 2 planning systems for all gamma criteria combined with all data grouped together (clinical and 3DTPS test) (see Table 3). Discussion At the time of implementing this audit, several institutions were requesting to use RIMRT in clinical trials and hence the audit was organised in collaboration between the National Cancer Research Institute Radiotherapy Trials Quality Assurance group (NCRI RTTQA), the Institute of Physics and Engineering in Medicine (IPEM) and the UK primary standards laboratory (the National Physical Laboratory (NPL)) to provide a national audit as well as credential institutions for RIMRT in clinical trials. Data collected

275

C.H. Clark et al. / Radiotherapy and Oncology 113 (2014) 272–278

Fig. 2. 3%/3 mm gamma passing rates for the three measured planes in the 3DTPS plan for each institution (where different equipment was used in a single institution i.e. VMAT and Tomotherapy systems, the data is shown as separate treatment planning and delivery combination).

Table 2 Median and range percentage pixels passing the gamma index criteria in all measured planes by TPS and delivery type for a range of gamma parameters for the 3DTPS and clinical plans. % Pixels passing

3%/3 mm

3%/2 mm

3DTPS

TPS (no.) Eclipse (22) HiArt (6) Monaco (6) OMP (5) Pinnacle (4) Delivery (no.) Elekta (12) Tomotherapy (6) Varian (25)

Clinical

2%/2 mm

3DTPS

Clinical

3DTPS

Clinical

Median

Range

Median

Range

Median

Range

Median

Range

Median

Range

Median

Range

100 100 99.6 97.0 95.6

96.9–100 91.6–100 89.7–100 83.8–100 84.0–100

100 100 100 99.3 98.8

97.2–100 98.4–100 95.3–100 94.8–100 94.3–100

99.7 100 96.8 93.0 90.6

90.5–100 85.0–100 76.9–100 67.6–100 67.0–98.8

99.8 99.7 99.5 96.0 96.6

92.0–100 96.2–100 81.7–100 85.2–98.2 88.9–99.8

98.2 98.5 92.6 85.0 82.0

78.6–100 70.7–100 66.8–99.7 50.7–97.8 50.5–92.0

97.3 97.2 98.8 85.4 88.7

79.9–100 91.2–99.7 70.2–100 62.5–95.0 78.3–96.1

97.7 100 100

83.8–100 91.6–100 84.0–100

99.7 100 100

94.3–100 98.4–100 97.2–100

93.6 100 99.7

67.6–100 85.0–100 67.0–100

96.6 99.7 99.7

81.7–100 96.2–100 92.0–100

84.5 98.5 97.9

50.7–99.7 70.7–100 50.5–100

89.1 97.2 96.7

62.5–100 91.2–99.7 79.9–100

during this audit has come from a wide range of treatment planning and delivery systems and therefore has the potential to benchmark what can be achieved with rotational IMRT systems and contribute data for use in setting tolerances for clinical trials and external audits. We have shown that for the 3DTPS test 79.1% of participating planning and delivery combinations achieved the standards recommended in the ESTRO booklet on QA in IMRT [2], the Institute of Physics and Engineering in Medicine (IPEM) Guidance for the Clinical Implementation of IMRT [17] and in the Code of Practice for QA and Control for IMRT published by the Netherlands Commission on Radiation Dosimetry (NCS) [18] of a pass rate higher than 95% using the 3%/3 mm gamma-index criterion. For the clinical plans this rose to 97.7% of planning and delivery combinations suggesting that the planning and delivery had been accurately implemented and perhaps that criteria could be tightened to 3%/2 mm (where 62.7% and 69.8% passed all planes >95% for the 3DTPS and clinical plans respectively, with 79% and 85% for Type 1) as appropriate gamma index parameters to aim for in RIMRT.

However this would mean that more centres would need to review their TPS model in order to join a clinical trial. Tolerances for inclusion in trials need to be chosen in a pragmatic manner to maintain standards, yet allow a sufficient number of centres to join for successful recruitment. The 3DTPS test has been found to be a challenging plan [13] and at least as hard to plan and verify as clinical trial plans (with breast, PPN and head and neck being increasingly challenging). For this reason we believe the 3DTPS test to be useful both for credentialing and commissioning (three centres revised their TPS model following the audit and subsequent analysis was carried out on the new model bringing the results to >95%). The majority of measured point doses were within ±3% of the TPS prediction, with the outliers mainly being in OARs where large percentage differences can be due to small absolute differences. Further, several of the outlying points in the OARs were where the TPS had overestimated the dose, which is less of an issue in terms of the clinical plan for the patient than if the TPS had underestimated. For the points in PTVs which were not within 3% we found that other points in the same

276

Rotational IMRT dosimetry audit

Fig. 3. Boxplots showing the gamma index pass rates for (a) the 3DTPS plans and (b) the clinical plans for each TPS and (c) the 3DTPS plans and (d) the clinical plans for each delivery system; for gamma index criteria of 3%/3 mm, 3%/2 mm and 2%/2 mm with interquartile range, median, minimum and maximum shown. (The stars and circles are counted as outliers. A case is identified as an outlier (circle) if its value is less than or equal to the first quartile minus 1.5 times the interquartile range, or is greater than or equal to the third quartile plus 1.5 times the interquartile range. If the case has a value less than or equal to the first quartile minus 3 times the interquartile range or greater than the third quartile plus 3 times the interquartile range, it is characterised as a far outlier (stars). When the cases are counted as outliers or far outliers, then they would not be counted as the minimum or maximum of the data range).

Table 3 Gamma index passing rate data by TPS type for all measured planes. TPS

Type 1 Type 2

% Pixels passing 3%/3 mm

% Pixels passing 3%/2 mm

% Pixels passing 2%/2 mm

Median

Range

Median

Range

Median

Range

100 98.8

91.6–100 83.8–100

99.8 96.2

85.0–100 67.0–100

97.7 87.4

70.7–100 50.5–100

or other PTVs in the plan were within 3% which was commonly due to the priorities that the planner had placed on the different PTVs and hence sometimes allowing gradients in the lower prioritised PTVs. The process of checking the plans prior to the audit visit led to the auditing group having a significant level of involvement with the institutions before the measurements were made. This meant that there was a high level of understanding at the institutions of what the audit requirements were and this may have had an impact on the overall results. Other audits which have been carried out by post [7,19,20] may have had more issues with a lack of

understanding of the protocol or human error which could have affected the results of the audit. Our data has shown a range of typical pass rates for given plans which vary between TPS. The statistically significant difference between the Type 1 and Type 2 TPS suggests that it is easier to achieve good results from Type 1 systems (in agreement with data from Molineu et al. [7]) where the TPS model has been designed specifically for a given delivery system and that the commissioning procedure may be better defined by the vendor. Further inspection of the Type 2 TPS results suggests that even when the beam modelling is performed by the TPS vendor rather than the user (as for

C.H. Clark et al. / Radiotherapy and Oncology 113 (2014) 272–278

Monaco where the results are excellent in some centres), there is still an effect of the TPS being designed to model multiple linac vendors. Furthermore not all Type 2 TPS allow for couch modelling. Further investigations with systems which do allow some form of couch modelling have suggested that this can make a difference of up to 2% in points measured in a centrally located PTV [21]. A recent AAPM 176 publication recommends that ‘the TPS should have models of the common couch tops from all vendors in the software’ [22]. Issues were also identified relating to use of too small a minimum leaf gap. This can allow creation of highly modulated fields and highly conformed dose distributions, but also has a tendency to use higher MUs, generally leading to less good verification measurements due to a less accurate TPS model of the very small segments. A further point is that the linacs without continuously variable dose rate capability had less smooth gantry rotation and this led to lower pass rates in the verification measurements. Overall there is a general lack of information/knowledge as to what some TPS/Linac combinations (in particular the Type 2) are capable of achieving. We have found that it is possible to get excellent results from both types of TPS and from all delivery systems, given sufficient commissioning; hence it would potentially aid users if more information were available on how to get the best from their TPS. Limitations of this audit have been that the phantom used (Octavius II) was a homogeneous phantom and therefore no assessment can be made of the system’s ability to correctly model RIMRT fields in heterogeneous tissues. Only 2D planar information has been assessed with a 2D gamma calculation; however the use of coronal and sagittal orientations of the array meant that verification was made in 3 dimensions within the plans. More recent commercial developments of this phantom and detector combination now offer 3D measurements and gamma calculation. However, although 3D gamma pass rates appear to be higher than 2D pass rates [23] it is not certain that the 3D approach would identify errors which may be missed by a 2D approach. In this study we have maintained classical analysis and reporting techniques as to our knowledge this is the first time an array has been used for an RIMRT audit. Future work is needed to look at how best to analyse the data in detail and whether to use single gamma parameters for the entire plan or whether tailoring gamma to the volume of interest might be more appropriate (e.g. low % dose difference and low distance to agreement for PTVs and high % dose difference and low distance to agreement for OARs). This audit was carried out as a one off exercise; however a follow up survey to ascertain the value of this audit to the institutions has indicated 95% of respondents have said that there should be another audit of this kind in 2–3 years’ time and 86% of respondents indicated that they would consider paying for it. Conclusions A dosimetry audit of rotational IMRT has been undertaken and has shown that TPS modelling and treatment delivery can achieve high gamma pass rates, including those recommended by national and international IMRT guidelines. Whilst statistically significant difference was seen between TPS where modelling had been specifically designed for the manufacturer’s own delivery system and TPS which had been designed to be independent of delivery system, overall the data confirmed that accurate implementation and delivery of rotational IMRT, suitable for use in clinical trials, is achievable. Conflicts of interest None declared.

277

Acknowledgements We would like to thank the staff from the following hospitals for their hard work in preparation for this audit and their kind hosting: Addenbrooke’s Hospital, Cambridge; Beatson Hospital, Glasgow; Castle Hill Hospital, Hull; Christie Hospital, Manchester; Clatterbridge Cancer Centre, Liverpool; Derriford Hospital, Plymouth; Guys and St Thomas’, London; Harley St Clinic, London; Ipswich Hospital, Ipswich; James Cook University Hospital, Middlesbrough; Kent Oncology Centre, Canterbury and Maidstone; Leicester Royal Infirmary, Leicester; Norfolk and Norwich University Hospital, Norwich; Northern Centre for Cancer Care, Newcastle; Northern Ireland Cancer Centre, Belfast; Peterborough City Hospital, Peterborough; Queen Elizabeth Hospital, Birmingham; Raigmore Hospital, Inverness; Royal Bath, Hospital, Bath; Royal Cornwall Hospital, Truro; Royal Devon and Exeter Hospital, Exeter; Royal Marsden Hospital, London; Royal Marsden Hospital, Sutton; Royal Surrey County Hospital, Guildford; Southampton General Hospital, Southampton; Southend Hospital, Southend; St Bartholomew’s Hospital, London; St James’ Institute for Oncology, Leeds; University College London Hospital, London; University Hospital North Staffordshire, Stoke; University Hospital, Nottingham; Velindre Cancer Centre, Cardiff; Western General Hospital, Edinburgh; Weston Park Hospital, Sheffield. We would also like to acknowledge Elizabeth Miles for her continued support and the funding from the National Measurement Systems and the National Cancer Research Institute. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.radonc.2014. 11.015. References [1] Teoh M, Clark CH, Wood K, Whitaker S, Nisbet A. Volumetric modulated arc therapy: a review of current literature and clinical use in practice. Br J Radiol 2011;84:6. [2] ESTRO 2008 Guidelines for the verification of IMRT http://www.estro.be/ ESTRO/upload/publications/Bookletn9_P3.pdf [3] Gillis S, De Wagter C, Bohsung J, Perrin B, Williams P, Mijnheer BJ. An intercentre quality assurance network for IMRT verification: preliminary results of the European QUASIMODO project. Radiother Oncol 2005;6:340–53. [4] Ibbott GS, Molineu A, Followill DS. Independent evaluations of IMRT through the use of an anthropomorphic phantom. Technol Cancer Res Treat 2006;5:481–7. [5] Budgell G, Berresford J, Trainer M, Bradshaw E, Sharpe P, Williams P. A national dosimetric audit of IMRT. Radiother Oncol 2011;99:2. [6] Schiefer H, Fogliata A, Nicolini G, Cozzi L, Seelentag WW, Born E, et al. The Swiss IMRT dosimetry intercomparison using a thorax phantom. Med Phys 2010;37:4424. [7] Molineu A, Hernandez N, Nguyen T, Ibbott G, Followill D. Credentialing results from IMRT irradiations of an anthropomorphic head and neck phantom. Med Phys 2013;40:022101. [8] Clark CH, Hansen VN, Chantler H, et al. Dosimetry audit for a multi-centre IMRT head and neck trial. Radiother Oncol 2009;93:102–8. [9] Ezzell G, Burmeister J, Dogan N, et al. IMRT commissioning: multiple institution planning and dosimetry comparisons, a report from AAPM Task Group 119. Med Phys 2009;36:5359–73. [10] Hussein M, Tsang Y, Thomas RAS, et al. A methodology for dosimetry audit of rotational radiotherapy using a commercial detector array. Radiother Oncol 2013;108:78–85. [11] Hussein M, Adams EJ, Jordan TJ, Clark CH, Nisbet A. Characterization of the sensitivity and resolution of a commercial 2D detector array for IMRT and VMAT verification. J App Clin Med Phys 2013;14:4460. [12] Hussein M, Rowshanfarzad P, Ebert MA, Nisbet A, Clark CH. A comparison of the gamma index analysis in various commercial IMRT/VMAT QA systems. Radiother Oncol 2013;109:370–6. [13] Tsang Y, Ciurlionis L, Clark CH, Venables K. Development of a novel treatment planning module for credentialing rotational IMRT techniques in the UK. Br J Radiol 2013;86:20120315. [14] National Cancer Research Institute Radiotherapy Trials Quality Assurance group (NCRI RTTQA) website: http://www.rttrialsqa.org.uk [accessed 29 August 2014].

278

Rotational IMRT dosimetry audit

[15] VODCA. Visualisation & Organisation of Data for Cancer Analysis. Switzerland: MSS Medical Software Solution GmbH; c2008 – 2012. Available from: http:// www.vodca.ch [accessed 9 October 2012]. [16] Low DA, Harms WB, Mutic S, Purdy JA. A technique for the quantitative evaluation of dose distributions. Med Phys 1998;25:656–61. [17] IPEM IMRT Working Party. Guidance for the Clinical Implementation of Intensity Modulated Radiation Therapy. IPEM Report 96. 2008. [18] Netherlands Commission on Radiation Dosimetry Subcommittee. Code of Practice for the Quality Assurance and Control for Intensity Modulated Radiotherapy. NCS report 22. 2013. [19] Izewska J, Georg D, Bera P, et al. A methodology for TLD postal dosimetry audit of high-energy radiotherapy photon beams in non-reference conditions. Radiother Oncol 2007;84:67–74.

[20] Williams I, Kenny J, Lye J, Lehmann J, Dunn L, Kron T. The Australian clinical dosimetry service: a commentary on the first 18 months. Australas Phys Eng Sci Med 2012;35:407–11. [21] Pulliam KB, Howell RM, Followill D, Luo D, White RA, Kry SF. The clinical impact of the couch top and rails on IMRT and arc therapy. Phys Med Biol 2011;56:7435–47. [22] Olch AJ, Gerig L, Heng L, Mihaylov I, Morgan A. Dosimetric effects caused by couch tops and immobilization devices: report of AAPM Task Group 176. Med Phys 2014;41:061501. [23] Pulliam KB, Huang JY, Howell RM. Comparison of 2D and 3D gamma analyses. Med Phys 2014;41:021710.