Independent review of 4DCT scans used for SABR treatment planning

Independent review of 4DCT scans used for SABR treatment planning MacCallum Cancer Centre, Melbourne, Vic, Australia Four‐dimensional computerized tomography (4DCT) is required for stereotactic abla- Centre for Medical Radiation Physics, tive body radiotherapy (SABR) of mobile targets to account for tumor motion during University of Wollongong, Wollongong, NSW, Australia treatment planning and delivery. In this study, we report on the impact of an image Department of Radiation Oncology, Peter review quality assurance process performed prior to treatment planning by medical MacCallum Cancer Centre, Melbourne, Vic, physicists for 4DCT scans used for SABR treatment. Reviews were performed of Australia Sir Peter MacCallum Department of 211 4DCT scans (193 patients) over a 3‐yr period (October 2014 to October 2017). Oncology, University of Melbourne, Treatment sites included lung (n = 168), kidney/adrenal/adrenal gland (n = 12), rib Melbourne, Vic, Australia (n = 4), mediastinum (n = 10), liver (n = 2), T‐spine (n = 1), and other abdominal sites (n = 14). It was found that in 23% (n = 49) of cases patient management was Author to whom correspondence should be altered due to the review process. The most frequent intervention involved patient‐ addressed. Peta Lonski E‐mail : Peta.Lonski@petermac.org specific contouring advice (n = 35 cases, 17%) including adjustment of internal tar- get volume (ITV) margins. In 13 cases (6%) a rescan was requested due to extensive motion artifact rendering the scan inadequate for SABR treatment planning. 4DCT review by medical physicists was found to be an effective method to improve plan quality for SABR. KEY W ORD S 4DCT, motion management, radiotherapy, SABR 1 | INTRODUCTION the breathing cycle. From the tumor motion in the individual phases, one can generate an internal target volume (ITV) which encompasses Stereotactic ablative body radiotherapy (SABR) is characterized by the GTV as well as its motion. Due to the risk of artifacts in 4DCTs, high radiation doses delivered in one or few treatment fractions. our institution has adopted a policy that these scans are reviewed SABR has been shown to be safe and effective for patients with by a medical physicist prior to treatment planning to ensure that the 1–3 4,5 early‐stage non‐small cell lung cancer and kidney cancer, and image is suitable for approximation of the tumor motion due to res- 6,7 8 9,10 shown promise for liver, spine and oligometastatic disease, as piration as well as for the creation of a reasonable reference image well as pancreas and prostate, in select patients. SABR treatment for image guidance. Ideally, the ITV contour must encompass the requires image guidance for accurate delivery, particularly for mobile size of the tumor as well as its full excursion throughout the entire targets. Patient immobilization and motion management strategies respiratory cycle. are used to ensure treatment is delivered as planned. For mobile tar- Irregular breathing patterns or 4DCT reconstruction errors such gets, retrospectively binned 4D‐computed tomography (4DCT) scans as one phase not reconstructing properly may lead to systematic may be performed to generate volumetric images at each phase of errors in ITV delineation propagating through the treatment chain -- -- --- -- --- -- --- -- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- -- --- -- --- -- --- -- --- -- -- - - - --- -- --- -- -- --- -- --- -- --- -- --- -- --- - This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, Inc. on behalf of American Association of Physicists in Medicine. 62 wileyonlinelibrary.com/journal/jacmp J Appl Clin Med Phys 2020; 21:3:62–67 ANTONY ET AL. | 63 which may not be obvious upon reviewing the maximum intensity used however if irregular breathing was noted during the surview projection (MIP) or average scans alone. Recently the European scan, radiation therapists would pause the scan procedure to provide Organisation for Research and Treatment of Cancer (EORTC) multi- basic coaching, although some patients still could not breath regu- centre Lungtech trial reported on the results of RTQA activities for larly throughout the entire scan. The resulting respiratory trace was 4DCT across 11 centers. Large deviations in contour volume of up used for phase binning, creating 10 phases of the breathing cycle. to 99% were found across different sites despite imaging the same Maximum intensity projection (MIP) and average datasets, which are phantom under the same motion pattern. The effects of irregular used for ITV delineation and dose calculation, respectively, were breathing patterns on ITV delineation of moving targets in the con- reconstructed from the raw data. The MIP was used for ITV delin- text of lung SABR have been described by Clements et al. who eation. 4DCT scans were reviewed by a medical physicist prior to demonstrated erroneous ITV delineation using MIP images for mov- treatment planning. Review was performed on the CT console, using ing targets with large amplitude undergoing irregular motion pat- the PulmoViewer application. This application allows visualization of terns. Similar findings have been reported by Park et al. who the breathing trace with the 4DCT image data, as well as tools to determined that the MIP underestimated the true target motion in determine the corresponding breath at each superior‐inferior scan the case of irregular motion. Measured PTV dose discrepancies of location. Tumor motion was measured using the ruler tool provided greater than 10% were reported by Huang et al. for irregular in PulmoViewer to assess the maximum displacement of the lesion motion patterns in a moving phantom for targets with large excur- between maximum inhale and exhale phases. The tumor boundaries sions, demonstrating systematic under‐dose of the PTV periphery in were identified using the radiation oncologists contour when avail- such cases. Clinical consequences may be severe, since systematic able, or through diagnostic imaging in consultation with radiation PTV under‐dosing from inappropriate ITV delineation will result in oncologists. A single, well‐defined edge of the tumor on each phase compromised tumor control probability. This risk is heightened in the was used to determine motion, therefore this is an estimate of superior‐inferior direction for co‐planar field deliveries where the tumor motion rather than the motion of the centre of mass. Choice dose falloff is steepest. Therefore, it is essential that the appropriate- of tumor edge was at the discretion of the reviewing physicist and ness of all 4DCT imaging be verified prior to clinical use to ensure was case‐specific, though usually the inferior‐most aspect of the that images derived are a true representation of the full tumor lesion was chosen if well‐defined. If the 4DCT was deemed by the excursion, particularly in cases of irregular breathing. medical physicist to not be an accurate representation of tumor This study presents the findings of independent, prospective motion, advice was provided on whether to rescan the patient or reviews performed by radiation oncology medical physicists of 211 adjust planning target volume (PTV) margins to account for increased patient 4DCT scans acquired for SABR pretreatment planning in a uncertainty, along with an estimate of the uncertainty. large radiotherapy centre. We report on the frequency of required 4D‐computed tomography review data were collected from three intervention as a result of the review process and correlation with radiotherapy facilities across our institution over a 3‐yr period regularity of patient breathing trace. between October 2014 and October 2017. Outcomes of the 4DCT reviews were assessed and each patient breathing trace were classi- fied according to regularity. Respiration cycles were classified as 2 | MATERIALS AND METHODS either “regular”, “adequate”,or “irregular”. For a breathing trace to be considered “regular”, the breathing pattern had to be consistent, Review guidelines for 4DCT image sets were developed based on repetitive in its amplitude and frequency, and free of significant commissioning work and experiences from quality assurance for irregularities, such as a halt in breathing or considerable change in 17,18 several clinical trials. An in‐house training programme was devel- breathing pattern. “Adequate” scans contained some irregularities, oped for medical physicists to establish a minimum skillset for per- such as a change in breathing pattern, but not affecting the tumor forming 4DCT reviews in the context of SABR. A patient‐specific level. “Irregular” scans contained considerable irregularity in breath- review checklist was designed to aid in the review process and facili- ing pattern at some point during scanning level of tumor excursion, tate data collection, which has been provided as supplementary or a change in breathing at the tumor level severe enough such that material. the subsequent image would not fully capture the tumor motion. 4D‐computed tomography scans were acquired on a Brilliance Examples of “irregular” breathing traces at the tumor level are shown widebore 16‐slice scanner (Philips Medical Systems, Eindhoven, the in Fig. 1. Breathing classification was made qualitatively, based on Netherlands) using retrospective gating with a gantry rotation period the judgment of the reviewing medical physicist. Additionally, the of 0.44 s, 140 kVp and a pitch adjusted based on the breathing rate tumor size and motion was documented for each case, including with a resulting patient dose approximately twice the one of a 3D whether hysteresis was evident in the tumor excursion throughout scan. 4DCT was also performed for lesions where dose calculation the respiratory cycle. Hysteresis was determined by observation of was likely to be affected by surrounding mobile structures, such as tumor motion on all phases viewed on the sagittal plane. Tumor ribs and lower thoracic spine at the level of the diaphragm. Respira- motion in the anterior‐posterior direction as well as superior‐inferior tion was monitored using the Philips bellows system affixed to the was classified as containing hysteresis. Reported CT dose index patients’ abdomen. Audio or visual coaching was not routinely (CTDI), pitch and breathing rates were also recorded. Breathing rates ANTONY ET AL. (a) (b) F IG.1. Examples showing irregular breathing in the case of (a) breathing stopped during scanning at the tumor level, and (b) irregular breath at the tumor level despite otherwise regular breathing. The cross‐hairs indicate that the tumor and the arrows mark the breathing track at the tumor level were measured using the tool provided in the PulmoView software, which reports both the average and location‐specific breathing rate as chosen by the user. The outcomes from each review regarding no issues (n=162) patient management were also assessed. 100 advice on margins or contouring (n=36) re-scan (n=13) 3 RESULTS Between October 2014 and October 2017, a total of 597 patients scanned with 4DCT were treated using SABR. Of those, 211 4DCT scan records were available for this retrospective audit. Target loca- tions included lung (n = 168), kidney/adrenal/adrenal gland (n = 12), regular adequate irregular rib (n = 4), mediastinum (n = 10), liver (n = 2), T‐spine (n = 1) and breathing trace classification other abdominal sites (n = 14). Review of 4DCT scans required approximately 20 min of medical physicists’ time per patient. As the FIG.2. Distribution of patient breathing traces according to respiratory cycle regularity for 211 SABR patients. Change in patient SABR programme increased capacity, the number of 4DCT reviews management as a result of 4DCT review is indicated by the shaded was found to steadily increase. bars. A total of 49 cases (23%) required change in patient The number of patient breathing traces which were considered management. Of those, 25 (51%) were classified as ‘adequate’ or “regular”, “adequate” or “irregular” is shown in Fig. 2. The impact on ‘irregular’ breathing patient management for each category is also shown. No issues were found for 162 patients (77%) and the scans were used for breathing rate throughout the scan did not necessarily predict inter- SABR treatment planning without intervention. Of those 162 vention requirements. patients, 136 (84%) had regular breathing patterns, 19 (12%) had The amplitude of total tumor motion is shown in Fig. 4 as a adequate regularity and 7 (4%) were considered irregular. For function of breathing rate at the tumor level, with data grouped remaining cases (n = 49, 23%), 4DCT reviews revealed issues with according to intervention requirements. Tumors with motion less the final images and required intervention. A re‐scan was subse- than 3 mm did not require intervention regardless of breathing rate. quently requested in 13 cases (6%) due to excessive motion artifact Large tumor excursion or rapid breathing rate were not predictors rendering the final images unsuitable for ITV delineation for SABR for intervention. treatment planning. For remaining cases (n = 35, 17%), advice to use Table 1 shows the frequency of tumor hysteresis throughout the modified margins in ITV delineation or other contouring advice respiratory cycle. Hysteresis was observed in 30% of patients in this including fusion of staging images such as PET was provided to com- study and is often noted for inferiorly located lesions close to the pensate for deficiencies in the 4DCT scan based on advice from the posterior chest wall. reviewing medical physicist. Comments in the review form were reviewed to determine the Figure 3 shows the average breathing rate throughout the 4DCT cause of the artifacts. A number of common causes were identified: scan and recorded breathing rate at the tumor level, with data 1. The patient’s breathing was highly irregular, leading to poor grouped according to intervention type. The line of identity is shown tumor definition in any one phase, and insufficient quality to by a solid line with a ±10% margin indicated by the dashed lines. It determine range of tumor motion. can be seen that breathing rate at the tumor level compared to number of patients ANTONY ET AL. | 65 35 TAB LE 1 Summary cases involving tumor hysteresis. Hysteresis no issues was observed in 64 out of 211 4DCT scans (30%). For the 48 cases advice on margin or contouring requiring some change in patient management, 23 cases (48%) were re-scan observed to have tumor hysteresis compared with 41 out of 163 (25%) of cases where no intervention was required Re‐scan or No issues advice Hysteresis n% n % Yes/Slight 41 25 23 48 No 122 75 25 52 5 expiration (inspiration) not being recorded, i.e., lack of informa- 515 25 35 tion on either end of the tumor excursion. average breathing rate (BPM) 6. The patient had an unintended deep inspiration while the tumor was moving through the scanning plane, leading to overestima- F IG.3. Correlation between average breathing rate throughout the 4DCT scan duration and breathing rate at the tumor level. The tion of tumor motion solid line represents the line of identity and the dashed lines represent ± 10% variation Reasons for physics consultations other than due to breathing irregularity and motion estimation included overestimation of the required tube current by the scanner software, slow breathing pat- no issues terns (<10 bpm not allowing 4DCT acquisition), inaccurate detection advice on margins or contouring of inhale peaks by the scanner software and poor image quality. re-scan 4 DISCUSSION This study reports on the outcomes of independent review for patient 4DCT scans acquired for treatment of SABR to mobile tar- gets. The aim of these reviews was to determine if each scan was a reasonable representation of tumor motion throughout the breathing cycle and was appropriate for the purposes of SABR treatment plan- ning, including target (ITV) delineation and dose calculation. 010 20 30 40 One limitation of this study is the subjectiveness amongst differ- total tumour amplitude [mm] ent physicists in performing quantitative analysis of patient 4DCT F IG.4. Change in patient management is shown relative to tumor reviews. While training was provided to harmonize interpretation, amplitude and patient breathing rate (breaths per minute, BPM) at there is still a degree of subjectiveness in the review process. Never- the tumor level. Motion less than 3 mm required no intervention. theless, intervention was required in 23% of all reviewed cases. Breathing rate was not a predictor for intervention requirements Irregular breathing rate was found to be a contributor to inadequate scans (16% of regular breathing traces requiring intervention com- 2. The patient was breathing regularly, but coughed during the pared to 57% of scans classified as “irregular”, Fig. 2). One common acquisition problem was identified as inappropriate choice of scan pitch. Scan- 3. Patient was breathing regularly, but while the tumor was moving ner pitch is adjusted based on patient breathing rate prior to com- through the scanning plane the patient stopped breathing, lead- mencing a scan. A lower pitch is required to maximize the chance of ing to the tumor appearing artificially stationary, with anatomy fully capturing tumor motion in the case of slower breathing rates. superior and inferior moving with respiration. The pitch is selected after the patient has spent some time in quiet 4. Patient’s breathing continuously slowed down from initial scan breathing and is monitored up until commencing a scan. However, pitch setting to acquisition. This may have been due to medica- upon commencing a scan it was found in some cases that a patient tion to relax the patient for the scan breathing rate can change, even throughout the duration of the scan. 5. The patient did not exhale (or inhale) fully, while scanning In some cases the breathing stopped completely while scanning through the superior (or inferior) aspect of the tumor. This through the level of the lesion, resulting in no visible tumor motion. resulted in the superior (or inferior) aspect of the tumor at full In such cases a rescan is required which usually addressed concerns breathing rate at tumour level (BPM) breathing rate at tumour level (BPM) ANTONY ET AL. raised in the first scan, unless a similar interruption in breathing pat- ITVs based on respiratory‐gated 4DCT are therefore necessary for tern occurred. In some cases irregular breathing was noted during improving target definition. The additional anterior‐posterior and the scan but no intervention was required. This may be due to the left‐right motion requires careful consideration of each phase of the irregularity occurring at anatomical locations away from the target breathing cycle, since the maximum inhale and maximum exhale may region. In such cases, irregular breathing is noted but if the target not capture the intermediate motion patterns. Use of the maximum region is unaffected intervention is not warranted. Since changes in intensity projection (MIP) image or all individual phases for ITV delin- breathing rate were shown to be a significant contributor to motion eation ensures tumors with hysteresis are fully captured. artifacts in our centre, radiation therapists have subsequently begun monitoring the respiratory trace closely during a scan. If irregular breathing is indicated during a scan, a physicist is called to review 5 | CONCLUSIONS the respiratory trace while the patient is still on‐site. This facilitates more timely re‐scans where warranted without the need to call a Patient‐specific 4DCT reviews by a medical physicist was shown to patient back to hospital. have a significant impact on patient management in a large cohort of Figure 1 shows that both large [Fig. 1(a)] and quite subtle patients treated with SABR to moving lesions with a high interven- [Fig. 1(b)] irregularities can impact on motion assessment. Both tion rate of 23% of all cases. Irregular breathing patterns during breathing frequency and amplitude can have a detrimental impact. 4DCT scans were shown to cause artefacts which may impact on Through the examples shown in this study, amplitude can have a the resulting ITV contours, hence treatment fields. In 23% of cases major impact if the tumor isn’t moving its “normal” extent during the physicist was able to advise on margins to accommodate for lost acquisition then tumor motion will not be sufficiently captured. motion during the scan, while in other cases a rescan was required. However, irregularities in frequency also impact our assessment due Tumor hysteresis was noted in 30% of scans, requiring careful to discontinuity artifacts, which is often inter‐related to image acqui- review of all phases to ensure tumor excursion is fully captured in all sition parameters such as pitch factor and gantry speed. It is thus directions of motion. Results from this study suggest patient‐specific quite challenging to quantify respiratory trace irregularities in a man- 4DCT QA should be a mandatory part of a patient’s treatment path- ner that can be applied routinely in the clinic. Thus, ongoing patient‐ way in SABR treatments of moving targets to ensure motion is ade- specific reviews are required. quately captured for the purposes of motion management and Typically a 4DCT scan acquires images of each anatomical slice treatment planning. for the duration of one to two breaths. Just one irregular breath can therefore distort the resulting image at a given anatomical slice. ACKNOWLEDGMENTS Review of PET scans (if available) acquired over several minutes was used to augment the relevant information where necessary. Also the Shankar Siva, Tomas Kron and Nicholas Hardcastle receive funding CBCT, or 4D‐CBCT if available, on the first treatment day can be from Varian Medical Systems for an unrelated project. used to validate the motion estimates. 4D cone‐beam CTs were occasionally acquired to evaluate motion, as these are more robust to breathing irregularity due to the whole anatomy being imaged for CONFLICT OF INTERESTS at least 2 min worth of breathing. It should be noted that due to the The author have no relevant conflict of interest to disclose. fact that 4DCTs are only acquiring motion from 1 to 2 breaths, cou- pled with the sampling frequency, the treatment respiratory motion REFERENCES is underestimated in 4DCTs. This means that any underestimation of the motion from 4DCTs is potentially more significant relative to 1. Timmerman R, Papiez L, McGarry R, et al. Extracranial stereotactic treatment motion. radioablation: results of a phase I study in medically inoperable stage I non‐small cell lung cancer. Chest. 2003;124:1946–55. Tumor hysteresis was noted in 30% of cases (n = 64). Of those, 2. Lagerwaard FJ, Haasbeek CJ, Smit EF, Slotman BJ, Senan S. Out- 48% required intervention compared to 25% of cases without hys- comes of risk‐adapted fractionated stereotactic radiotherapy for teresis. Although this study is not powered to compare intervention stage I non–small‐cell lung cancer. Int J Radiat Oncol Biol Phys. rates with and without tumor hysteresis the differences are worth 2008;70:685–92. 3. Ball D, Mai GT, Vinod S, et al. Stereotactic ablative radiotherapy ver- noting. It may be that a more complex motion pattern has a higher sus standard radiotherapy in stage 1 non‐small‐cell lung cancer chance of being missed in the presence of artifacts, compared to a (TROG 09.02 CHISEL): a phase 3, open‐label, randomised controlled more simple superior/inferior motion pattern. trial. Lancet Oncol. 2019;20:494–503. Earlier studies suggest that artifacts in 4DCT are common and 4. Pham D, Thompson A, Kron T, et al. Stereotactic ablative body radia- tion therapy for primary kidney cancer: a 3‐dimensional conformal associated with breathing irregularity. Patient training, coaching technique associated with low rates of early toxicity. Int J Radiat and feedback would be helpful to improve patient compliance with Oncol Biol Phys. 2014;90:1061–8. regular and reproducible breathing. Furthermore, thoracic lesions 5. Siva S, Pham D, Gill S, Corcoran NM, Foroudi F. A systematic review are subject to often complex motion patterns depending on the loca- of stereotactic radiotherapy ablation for primary renal cell carcinoma. BJU Int. 2012;110:E737–E743. tion and can even be affected by cardiac motion. Individualized ANTONY ET AL. | 67 6. Scorsetti M, Arcangeli S, Tozzi A, et al. Is stereotactic body radiation 18. Siva S, Kron T, Bressel M, et al. A randomised phase II trial of therapy an attractive option for unresectable liver metastases? A stereotactic ablative fractionated radiotherapy versus radiosurgery preliminary report from a phase 2 trial. Int J Radiat Oncol Biol Phys. for oligometastatic neoplasia to the lung (TROG 13.01 SAFRON II). 2013;86:336–42. BMC Cancer. 2016;16:183. 7. Fuss M, Thomas CR. Stereotactic body radiation therapy: an ablative 19. Hubbard P, Callahan J, Cramb J, Budd R, Kron T. Audit of radiation treatment option for primary and secondary liver tumors. Ann Surg dose delivered in time‐resolved four‐dimensional computed tomogra- Oncol. 2004;11:130–8. phy in a radiotherapy department. J Med Imaging Radiat Oncol. 8. Chang J, Gandhidasan S, Finnigan R, et al. Stereotactic ablative body 2015;59:346–52. radiotherapy for the treatment of spinal oligometastases. Clin Oncol 20. Glide‐Hurst CK, Smith MS, Ajlouni M, Chetty IJ. Evaluation of two (R Coll Radiol). 2017;29:e119–e25. synchronized external surrogates for 4D CT sorting. J Appl Clin Med 9. Alongi F, Arcangeli S, Filippi AR, Ricardi U, Scorsetti M. Review and Phys. 2013;14:117–32. uses of stereotactic body radiation therapy for oligometastases. 21. Steiner E, Shieh C‐C, Caillet V, et al. Both four‐dimensional com- Oncologist. 2012;17:1100–7. puted tomography and four‐dimensional cone beam computed 10. Hanna G, Landau D. Stereotactic body radiotherapy for oligometa- tomography under‐predict lung target motion during radiotherapy. static disease. Clin Oncol (R Coll Radiol). 2015;27:290–7. Radiother Oncol. 2019;135:65–73. 11. Chang BK, Timmerman RD. Stereotactic body radiation therapy: a 22. Yamamoto T, Langner U, Loo BW Jr, Shen J, Keall PJ. Retrospective comprehensive review. Am J Clin Oncol. 2007;30:637–44. analysis of artifacts in four‐dimensional CT images of 50 abdominal 12. Lambrecht M, Sonke J‐J, Nestle U, et al. Quality assurance of four‐ and thoracic radiotherapy patients. Int J Radiat Oncol Biol Phys. dimensional computed tomography in a multicentre trial of stereo- 2008;72:1250–8. tactic body radiotherapy of centrally located lung tumours. Phys 23. George R, Chung TD, Vedam SS, et al. Audio‐visual biofeedback for Imaging Radiat Oncol. 2018;8:57–62. respiratory‐gated radiotherapy: impact of audio instruction and 13. Clements N, Kron T, Franich R, et al. The effect of irregular breath- audio‐visual biofeedback on respiratory‐gated radiotherapy. Int J ing patterns on internal target volumes in four‐dimensional CT and Radiat Oncol Biol Phys. 2006;65:924–33. cone‐beam CT images in the context of stereotactic lung radiother- 24. Seppenwoolde Y, Shirato H, Kitamura K, et al. Precise and real‐time apy. Med Phys. 2013;40:021904. measurement of 3D tumor motion in lung due to breathing and 14. Park K, Huang L, Gagne H, Papiez L. Do maximum intensity projec- heartbeat, measured during radiotherapy. Int J Radiat Oncol Biol tion images truly capture tumor motion? Int J Radiat Oncol Biol Phys. Phys. 2002;53:822–34. 2009;73:618–25. 25. Underberg RW, Lagerwaard FJ, Cuijpers JP, Slotman BJ, De Koste 15. Huang L, Park K, Boike T, et al. A study on the dosimetric accuracy JRVS, Senan S. Four‐dimensional CT scans for treatment planning in of treatment planning for stereotactic body radiation therapy of lung stereotactic radiotherapy for stage I lung cancer. Int J Radiat Oncol cancer using average and maximum intensity projection images. Biol Phys. 2004;60:1283–90. Radiother Oncol. 2010;96:48–54. 16. Hardcastle N, Clements N, Chesson B, et al. Results of patient speci- fic quality assurance for patients undergoing stereotactic ablative SUPPORTING INFORMATION radiotherapy for lung lesions. Australas Phys Eng Sci Med. 2014;37:45–52. Additional supporting information may be found online in the 17. Kron T, Chesson B, Hardcastle N, et al. Credentialing of radiotherapy Supporting Information section at the end of the article. centres in Australasia for TROG 09.02 (Chisel), a Phase III clinical trial on stereotactic ablative body radiotherapy of early stage lung cancer. Br J Radiol. 2018;91:20170737. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Applied Clinical Medical Physics Pubmed Central

Independent review of 4DCT scans used for SABR treatment planning

Journal of Applied Clinical Medical Physics, Volume 21 (3) – Feb 13, 2020

Loading next page...
 
/lp/pubmed-central/independent-review-of-4dct-scans-used-for-sabr-treatment-planning-OkmCHT8MtE
Publisher
Pubmed Central
Copyright
© 2020 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, Inc. on behalf of American Association of Physicists in Medicine.
eISSN
1526-9914
DOI
10.1002/acm2.12825
Publisher site
See Article on Publisher Site

Abstract

MacCallum Cancer Centre, Melbourne, Vic, Australia Four‐dimensional computerized tomography (4DCT) is required for stereotactic abla- Centre for Medical Radiation Physics, tive body radiotherapy (SABR) of mobile targets to account for tumor motion during University of Wollongong, Wollongong, NSW, Australia treatment planning and delivery. In this study, we report on the impact of an image Department of Radiation Oncology, Peter review quality assurance process performed prior to treatment planning by medical MacCallum Cancer Centre, Melbourne, Vic, physicists for 4DCT scans used for SABR treatment. Reviews were performed of Australia Sir Peter MacCallum Department of 211 4DCT scans (193 patients) over a 3‐yr period (October 2014 to October 2017). Oncology, University of Melbourne, Treatment sites included lung (n = 168), kidney/adrenal/adrenal gland (n = 12), rib Melbourne, Vic, Australia (n = 4), mediastinum (n = 10), liver (n = 2), T‐spine (n = 1), and other abdominal sites (n = 14). It was found that in 23% (n = 49) of cases patient management was Author to whom correspondence should be altered due to the review process. The most frequent intervention involved patient‐ addressed. Peta Lonski E‐mail : Peta.Lonski@petermac.org specific contouring advice (n = 35 cases, 17%) including adjustment of internal tar- get volume (ITV) margins. In 13 cases (6%) a rescan was requested due to extensive motion artifact rendering the scan inadequate for SABR treatment planning. 4DCT review by medical physicists was found to be an effective method to improve plan quality for SABR. KEY W ORD S 4DCT, motion management, radiotherapy, SABR 1 | INTRODUCTION the breathing cycle. From the tumor motion in the individual phases, one can generate an internal target volume (ITV) which encompasses Stereotactic ablative body radiotherapy (SABR) is characterized by the GTV as well as its motion. Due to the risk of artifacts in 4DCTs, high radiation doses delivered in one or few treatment fractions. our institution has adopted a policy that these scans are reviewed SABR has been shown to be safe and effective for patients with by a medical physicist prior to treatment planning to ensure that the 1–3 4,5 early‐stage non‐small cell lung cancer and kidney cancer, and image is suitable for approximation of the tumor motion due to res- 6,7 8 9,10 shown promise for liver, spine and oligometastatic disease, as piration as well as for the creation of a reasonable reference image well as pancreas and prostate, in select patients. SABR treatment for image guidance. Ideally, the ITV contour must encompass the requires image guidance for accurate delivery, particularly for mobile size of the tumor as well as its full excursion throughout the entire targets. Patient immobilization and motion management strategies respiratory cycle. are used to ensure treatment is delivered as planned. For mobile tar- Irregular breathing patterns or 4DCT reconstruction errors such gets, retrospectively binned 4D‐computed tomography (4DCT) scans as one phase not reconstructing properly may lead to systematic may be performed to generate volumetric images at each phase of errors in ITV delineation propagating through the treatment chain -- -- --- -- --- -- --- -- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- --- -- -- --- -- --- -- --- -- --- -- -- - - - --- -- --- -- -- --- -- --- -- --- -- --- -- --- - This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, Inc. on behalf of American Association of Physicists in Medicine. 62 wileyonlinelibrary.com/journal/jacmp J Appl Clin Med Phys 2020; 21:3:62–67 ANTONY ET AL. | 63 which may not be obvious upon reviewing the maximum intensity used however if irregular breathing was noted during the surview projection (MIP) or average scans alone. Recently the European scan, radiation therapists would pause the scan procedure to provide Organisation for Research and Treatment of Cancer (EORTC) multi- basic coaching, although some patients still could not breath regu- centre Lungtech trial reported on the results of RTQA activities for larly throughout the entire scan. The resulting respiratory trace was 4DCT across 11 centers. Large deviations in contour volume of up used for phase binning, creating 10 phases of the breathing cycle. to 99% were found across different sites despite imaging the same Maximum intensity projection (MIP) and average datasets, which are phantom under the same motion pattern. The effects of irregular used for ITV delineation and dose calculation, respectively, were breathing patterns on ITV delineation of moving targets in the con- reconstructed from the raw data. The MIP was used for ITV delin- text of lung SABR have been described by Clements et al. who eation. 4DCT scans were reviewed by a medical physicist prior to demonstrated erroneous ITV delineation using MIP images for mov- treatment planning. Review was performed on the CT console, using ing targets with large amplitude undergoing irregular motion pat- the PulmoViewer application. This application allows visualization of terns. Similar findings have been reported by Park et al. who the breathing trace with the 4DCT image data, as well as tools to determined that the MIP underestimated the true target motion in determine the corresponding breath at each superior‐inferior scan the case of irregular motion. Measured PTV dose discrepancies of location. Tumor motion was measured using the ruler tool provided greater than 10% were reported by Huang et al. for irregular in PulmoViewer to assess the maximum displacement of the lesion motion patterns in a moving phantom for targets with large excur- between maximum inhale and exhale phases. The tumor boundaries sions, demonstrating systematic under‐dose of the PTV periphery in were identified using the radiation oncologists contour when avail- such cases. Clinical consequences may be severe, since systematic able, or through diagnostic imaging in consultation with radiation PTV under‐dosing from inappropriate ITV delineation will result in oncologists. A single, well‐defined edge of the tumor on each phase compromised tumor control probability. This risk is heightened in the was used to determine motion, therefore this is an estimate of superior‐inferior direction for co‐planar field deliveries where the tumor motion rather than the motion of the centre of mass. Choice dose falloff is steepest. Therefore, it is essential that the appropriate- of tumor edge was at the discretion of the reviewing physicist and ness of all 4DCT imaging be verified prior to clinical use to ensure was case‐specific, though usually the inferior‐most aspect of the that images derived are a true representation of the full tumor lesion was chosen if well‐defined. If the 4DCT was deemed by the excursion, particularly in cases of irregular breathing. medical physicist to not be an accurate representation of tumor This study presents the findings of independent, prospective motion, advice was provided on whether to rescan the patient or reviews performed by radiation oncology medical physicists of 211 adjust planning target volume (PTV) margins to account for increased patient 4DCT scans acquired for SABR pretreatment planning in a uncertainty, along with an estimate of the uncertainty. large radiotherapy centre. We report on the frequency of required 4D‐computed tomography review data were collected from three intervention as a result of the review process and correlation with radiotherapy facilities across our institution over a 3‐yr period regularity of patient breathing trace. between October 2014 and October 2017. Outcomes of the 4DCT reviews were assessed and each patient breathing trace were classi- fied according to regularity. Respiration cycles were classified as 2 | MATERIALS AND METHODS either “regular”, “adequate”,or “irregular”. For a breathing trace to be considered “regular”, the breathing pattern had to be consistent, Review guidelines for 4DCT image sets were developed based on repetitive in its amplitude and frequency, and free of significant commissioning work and experiences from quality assurance for irregularities, such as a halt in breathing or considerable change in 17,18 several clinical trials. An in‐house training programme was devel- breathing pattern. “Adequate” scans contained some irregularities, oped for medical physicists to establish a minimum skillset for per- such as a change in breathing pattern, but not affecting the tumor forming 4DCT reviews in the context of SABR. A patient‐specific level. “Irregular” scans contained considerable irregularity in breath- review checklist was designed to aid in the review process and facili- ing pattern at some point during scanning level of tumor excursion, tate data collection, which has been provided as supplementary or a change in breathing at the tumor level severe enough such that material. the subsequent image would not fully capture the tumor motion. 4D‐computed tomography scans were acquired on a Brilliance Examples of “irregular” breathing traces at the tumor level are shown widebore 16‐slice scanner (Philips Medical Systems, Eindhoven, the in Fig. 1. Breathing classification was made qualitatively, based on Netherlands) using retrospective gating with a gantry rotation period the judgment of the reviewing medical physicist. Additionally, the of 0.44 s, 140 kVp and a pitch adjusted based on the breathing rate tumor size and motion was documented for each case, including with a resulting patient dose approximately twice the one of a 3D whether hysteresis was evident in the tumor excursion throughout scan. 4DCT was also performed for lesions where dose calculation the respiratory cycle. Hysteresis was determined by observation of was likely to be affected by surrounding mobile structures, such as tumor motion on all phases viewed on the sagittal plane. Tumor ribs and lower thoracic spine at the level of the diaphragm. Respira- motion in the anterior‐posterior direction as well as superior‐inferior tion was monitored using the Philips bellows system affixed to the was classified as containing hysteresis. Reported CT dose index patients’ abdomen. Audio or visual coaching was not routinely (CTDI), pitch and breathing rates were also recorded. Breathing rates ANTONY ET AL. (a) (b) F IG.1. Examples showing irregular breathing in the case of (a) breathing stopped during scanning at the tumor level, and (b) irregular breath at the tumor level despite otherwise regular breathing. The cross‐hairs indicate that the tumor and the arrows mark the breathing track at the tumor level were measured using the tool provided in the PulmoView software, which reports both the average and location‐specific breathing rate as chosen by the user. The outcomes from each review regarding no issues (n=162) patient management were also assessed. 100 advice on margins or contouring (n=36) re-scan (n=13) 3 RESULTS Between October 2014 and October 2017, a total of 597 patients scanned with 4DCT were treated using SABR. Of those, 211 4DCT scan records were available for this retrospective audit. Target loca- tions included lung (n = 168), kidney/adrenal/adrenal gland (n = 12), regular adequate irregular rib (n = 4), mediastinum (n = 10), liver (n = 2), T‐spine (n = 1) and breathing trace classification other abdominal sites (n = 14). Review of 4DCT scans required approximately 20 min of medical physicists’ time per patient. As the FIG.2. Distribution of patient breathing traces according to respiratory cycle regularity for 211 SABR patients. Change in patient SABR programme increased capacity, the number of 4DCT reviews management as a result of 4DCT review is indicated by the shaded was found to steadily increase. bars. A total of 49 cases (23%) required change in patient The number of patient breathing traces which were considered management. Of those, 25 (51%) were classified as ‘adequate’ or “regular”, “adequate” or “irregular” is shown in Fig. 2. The impact on ‘irregular’ breathing patient management for each category is also shown. No issues were found for 162 patients (77%) and the scans were used for breathing rate throughout the scan did not necessarily predict inter- SABR treatment planning without intervention. Of those 162 vention requirements. patients, 136 (84%) had regular breathing patterns, 19 (12%) had The amplitude of total tumor motion is shown in Fig. 4 as a adequate regularity and 7 (4%) were considered irregular. For function of breathing rate at the tumor level, with data grouped remaining cases (n = 49, 23%), 4DCT reviews revealed issues with according to intervention requirements. Tumors with motion less the final images and required intervention. A re‐scan was subse- than 3 mm did not require intervention regardless of breathing rate. quently requested in 13 cases (6%) due to excessive motion artifact Large tumor excursion or rapid breathing rate were not predictors rendering the final images unsuitable for ITV delineation for SABR for intervention. treatment planning. For remaining cases (n = 35, 17%), advice to use Table 1 shows the frequency of tumor hysteresis throughout the modified margins in ITV delineation or other contouring advice respiratory cycle. Hysteresis was observed in 30% of patients in this including fusion of staging images such as PET was provided to com- study and is often noted for inferiorly located lesions close to the pensate for deficiencies in the 4DCT scan based on advice from the posterior chest wall. reviewing medical physicist. Comments in the review form were reviewed to determine the Figure 3 shows the average breathing rate throughout the 4DCT cause of the artifacts. A number of common causes were identified: scan and recorded breathing rate at the tumor level, with data 1. The patient’s breathing was highly irregular, leading to poor grouped according to intervention type. The line of identity is shown tumor definition in any one phase, and insufficient quality to by a solid line with a ±10% margin indicated by the dashed lines. It determine range of tumor motion. can be seen that breathing rate at the tumor level compared to number of patients ANTONY ET AL. | 65 35 TAB LE 1 Summary cases involving tumor hysteresis. Hysteresis no issues was observed in 64 out of 211 4DCT scans (30%). For the 48 cases advice on margin or contouring requiring some change in patient management, 23 cases (48%) were re-scan observed to have tumor hysteresis compared with 41 out of 163 (25%) of cases where no intervention was required Re‐scan or No issues advice Hysteresis n% n % Yes/Slight 41 25 23 48 No 122 75 25 52 5 expiration (inspiration) not being recorded, i.e., lack of informa- 515 25 35 tion on either end of the tumor excursion. average breathing rate (BPM) 6. The patient had an unintended deep inspiration while the tumor was moving through the scanning plane, leading to overestima- F IG.3. Correlation between average breathing rate throughout the 4DCT scan duration and breathing rate at the tumor level. The tion of tumor motion solid line represents the line of identity and the dashed lines represent ± 10% variation Reasons for physics consultations other than due to breathing irregularity and motion estimation included overestimation of the required tube current by the scanner software, slow breathing pat- no issues terns (<10 bpm not allowing 4DCT acquisition), inaccurate detection advice on margins or contouring of inhale peaks by the scanner software and poor image quality. re-scan 4 DISCUSSION This study reports on the outcomes of independent review for patient 4DCT scans acquired for treatment of SABR to mobile tar- gets. The aim of these reviews was to determine if each scan was a reasonable representation of tumor motion throughout the breathing cycle and was appropriate for the purposes of SABR treatment plan- ning, including target (ITV) delineation and dose calculation. 010 20 30 40 One limitation of this study is the subjectiveness amongst differ- total tumour amplitude [mm] ent physicists in performing quantitative analysis of patient 4DCT F IG.4. Change in patient management is shown relative to tumor reviews. While training was provided to harmonize interpretation, amplitude and patient breathing rate (breaths per minute, BPM) at there is still a degree of subjectiveness in the review process. Never- the tumor level. Motion less than 3 mm required no intervention. theless, intervention was required in 23% of all reviewed cases. Breathing rate was not a predictor for intervention requirements Irregular breathing rate was found to be a contributor to inadequate scans (16% of regular breathing traces requiring intervention com- 2. The patient was breathing regularly, but coughed during the pared to 57% of scans classified as “irregular”, Fig. 2). One common acquisition problem was identified as inappropriate choice of scan pitch. Scan- 3. Patient was breathing regularly, but while the tumor was moving ner pitch is adjusted based on patient breathing rate prior to com- through the scanning plane the patient stopped breathing, lead- mencing a scan. A lower pitch is required to maximize the chance of ing to the tumor appearing artificially stationary, with anatomy fully capturing tumor motion in the case of slower breathing rates. superior and inferior moving with respiration. The pitch is selected after the patient has spent some time in quiet 4. Patient’s breathing continuously slowed down from initial scan breathing and is monitored up until commencing a scan. However, pitch setting to acquisition. This may have been due to medica- upon commencing a scan it was found in some cases that a patient tion to relax the patient for the scan breathing rate can change, even throughout the duration of the scan. 5. The patient did not exhale (or inhale) fully, while scanning In some cases the breathing stopped completely while scanning through the superior (or inferior) aspect of the tumor. This through the level of the lesion, resulting in no visible tumor motion. resulted in the superior (or inferior) aspect of the tumor at full In such cases a rescan is required which usually addressed concerns breathing rate at tumour level (BPM) breathing rate at tumour level (BPM) ANTONY ET AL. raised in the first scan, unless a similar interruption in breathing pat- ITVs based on respiratory‐gated 4DCT are therefore necessary for tern occurred. In some cases irregular breathing was noted during improving target definition. The additional anterior‐posterior and the scan but no intervention was required. This may be due to the left‐right motion requires careful consideration of each phase of the irregularity occurring at anatomical locations away from the target breathing cycle, since the maximum inhale and maximum exhale may region. In such cases, irregular breathing is noted but if the target not capture the intermediate motion patterns. Use of the maximum region is unaffected intervention is not warranted. Since changes in intensity projection (MIP) image or all individual phases for ITV delin- breathing rate were shown to be a significant contributor to motion eation ensures tumors with hysteresis are fully captured. artifacts in our centre, radiation therapists have subsequently begun monitoring the respiratory trace closely during a scan. If irregular breathing is indicated during a scan, a physicist is called to review 5 | CONCLUSIONS the respiratory trace while the patient is still on‐site. This facilitates more timely re‐scans where warranted without the need to call a Patient‐specific 4DCT reviews by a medical physicist was shown to patient back to hospital. have a significant impact on patient management in a large cohort of Figure 1 shows that both large [Fig. 1(a)] and quite subtle patients treated with SABR to moving lesions with a high interven- [Fig. 1(b)] irregularities can impact on motion assessment. Both tion rate of 23% of all cases. Irregular breathing patterns during breathing frequency and amplitude can have a detrimental impact. 4DCT scans were shown to cause artefacts which may impact on Through the examples shown in this study, amplitude can have a the resulting ITV contours, hence treatment fields. In 23% of cases major impact if the tumor isn’t moving its “normal” extent during the physicist was able to advise on margins to accommodate for lost acquisition then tumor motion will not be sufficiently captured. motion during the scan, while in other cases a rescan was required. However, irregularities in frequency also impact our assessment due Tumor hysteresis was noted in 30% of scans, requiring careful to discontinuity artifacts, which is often inter‐related to image acqui- review of all phases to ensure tumor excursion is fully captured in all sition parameters such as pitch factor and gantry speed. It is thus directions of motion. Results from this study suggest patient‐specific quite challenging to quantify respiratory trace irregularities in a man- 4DCT QA should be a mandatory part of a patient’s treatment path- ner that can be applied routinely in the clinic. Thus, ongoing patient‐ way in SABR treatments of moving targets to ensure motion is ade- specific reviews are required. quately captured for the purposes of motion management and Typically a 4DCT scan acquires images of each anatomical slice treatment planning. for the duration of one to two breaths. Just one irregular breath can therefore distort the resulting image at a given anatomical slice. ACKNOWLEDGMENTS Review of PET scans (if available) acquired over several minutes was used to augment the relevant information where necessary. Also the Shankar Siva, Tomas Kron and Nicholas Hardcastle receive funding CBCT, or 4D‐CBCT if available, on the first treatment day can be from Varian Medical Systems for an unrelated project. used to validate the motion estimates. 4D cone‐beam CTs were occasionally acquired to evaluate motion, as these are more robust to breathing irregularity due to the whole anatomy being imaged for CONFLICT OF INTERESTS at least 2 min worth of breathing. It should be noted that due to the The author have no relevant conflict of interest to disclose. fact that 4DCTs are only acquiring motion from 1 to 2 breaths, cou- pled with the sampling frequency, the treatment respiratory motion REFERENCES is underestimated in 4DCTs. This means that any underestimation of the motion from 4DCTs is potentially more significant relative to 1. Timmerman R, Papiez L, McGarry R, et al. Extracranial stereotactic treatment motion. radioablation: results of a phase I study in medically inoperable stage I non‐small cell lung cancer. Chest. 2003;124:1946–55. Tumor hysteresis was noted in 30% of cases (n = 64). Of those, 2. Lagerwaard FJ, Haasbeek CJ, Smit EF, Slotman BJ, Senan S. Out- 48% required intervention compared to 25% of cases without hys- comes of risk‐adapted fractionated stereotactic radiotherapy for teresis. Although this study is not powered to compare intervention stage I non–small‐cell lung cancer. Int J Radiat Oncol Biol Phys. rates with and without tumor hysteresis the differences are worth 2008;70:685–92. 3. Ball D, Mai GT, Vinod S, et al. Stereotactic ablative radiotherapy ver- noting. It may be that a more complex motion pattern has a higher sus standard radiotherapy in stage 1 non‐small‐cell lung cancer chance of being missed in the presence of artifacts, compared to a (TROG 09.02 CHISEL): a phase 3, open‐label, randomised controlled more simple superior/inferior motion pattern. trial. Lancet Oncol. 2019;20:494–503. Earlier studies suggest that artifacts in 4DCT are common and 4. Pham D, Thompson A, Kron T, et al. Stereotactic ablative body radia- tion therapy for primary kidney cancer: a 3‐dimensional conformal associated with breathing irregularity. Patient training, coaching technique associated with low rates of early toxicity. Int J Radiat and feedback would be helpful to improve patient compliance with Oncol Biol Phys. 2014;90:1061–8. regular and reproducible breathing. Furthermore, thoracic lesions 5. Siva S, Pham D, Gill S, Corcoran NM, Foroudi F. A systematic review are subject to often complex motion patterns depending on the loca- of stereotactic radiotherapy ablation for primary renal cell carcinoma. BJU Int. 2012;110:E737–E743. tion and can even be affected by cardiac motion. Individualized ANTONY ET AL. | 67 6. Scorsetti M, Arcangeli S, Tozzi A, et al. Is stereotactic body radiation 18. Siva S, Kron T, Bressel M, et al. A randomised phase II trial of therapy an attractive option for unresectable liver metastases? A stereotactic ablative fractionated radiotherapy versus radiosurgery preliminary report from a phase 2 trial. Int J Radiat Oncol Biol Phys. for oligometastatic neoplasia to the lung (TROG 13.01 SAFRON II). 2013;86:336–42. BMC Cancer. 2016;16:183. 7. Fuss M, Thomas CR. Stereotactic body radiation therapy: an ablative 19. Hubbard P, Callahan J, Cramb J, Budd R, Kron T. Audit of radiation treatment option for primary and secondary liver tumors. Ann Surg dose delivered in time‐resolved four‐dimensional computed tomogra- Oncol. 2004;11:130–8. phy in a radiotherapy department. J Med Imaging Radiat Oncol. 8. Chang J, Gandhidasan S, Finnigan R, et al. Stereotactic ablative body 2015;59:346–52. radiotherapy for the treatment of spinal oligometastases. Clin Oncol 20. Glide‐Hurst CK, Smith MS, Ajlouni M, Chetty IJ. Evaluation of two (R Coll Radiol). 2017;29:e119–e25. synchronized external surrogates for 4D CT sorting. J Appl Clin Med 9. Alongi F, Arcangeli S, Filippi AR, Ricardi U, Scorsetti M. Review and Phys. 2013;14:117–32. uses of stereotactic body radiation therapy for oligometastases. 21. Steiner E, Shieh C‐C, Caillet V, et al. Both four‐dimensional com- Oncologist. 2012;17:1100–7. puted tomography and four‐dimensional cone beam computed 10. Hanna G, Landau D. Stereotactic body radiotherapy for oligometa- tomography under‐predict lung target motion during radiotherapy. static disease. Clin Oncol (R Coll Radiol). 2015;27:290–7. Radiother Oncol. 2019;135:65–73. 11. Chang BK, Timmerman RD. Stereotactic body radiation therapy: a 22. Yamamoto T, Langner U, Loo BW Jr, Shen J, Keall PJ. Retrospective comprehensive review. Am J Clin Oncol. 2007;30:637–44. analysis of artifacts in four‐dimensional CT images of 50 abdominal 12. Lambrecht M, Sonke J‐J, Nestle U, et al. Quality assurance of four‐ and thoracic radiotherapy patients. Int J Radiat Oncol Biol Phys. dimensional computed tomography in a multicentre trial of stereo- 2008;72:1250–8. tactic body radiotherapy of centrally located lung tumours. Phys 23. George R, Chung TD, Vedam SS, et al. Audio‐visual biofeedback for Imaging Radiat Oncol. 2018;8:57–62. respiratory‐gated radiotherapy: impact of audio instruction and 13. Clements N, Kron T, Franich R, et al. The effect of irregular breath- audio‐visual biofeedback on respiratory‐gated radiotherapy. Int J ing patterns on internal target volumes in four‐dimensional CT and Radiat Oncol Biol Phys. 2006;65:924–33. cone‐beam CT images in the context of stereotactic lung radiother- 24. Seppenwoolde Y, Shirato H, Kitamura K, et al. Precise and real‐time apy. Med Phys. 2013;40:021904. measurement of 3D tumor motion in lung due to breathing and 14. Park K, Huang L, Gagne H, Papiez L. Do maximum intensity projec- heartbeat, measured during radiotherapy. Int J Radiat Oncol Biol tion images truly capture tumor motion? Int J Radiat Oncol Biol Phys. Phys. 2002;53:822–34. 2009;73:618–25. 25. Underberg RW, Lagerwaard FJ, Cuijpers JP, Slotman BJ, De Koste 15. Huang L, Park K, Boike T, et al. A study on the dosimetric accuracy JRVS, Senan S. Four‐dimensional CT scans for treatment planning in of treatment planning for stereotactic body radiation therapy of lung stereotactic radiotherapy for stage I lung cancer. Int J Radiat Oncol cancer using average and maximum intensity projection images. Biol Phys. 2004;60:1283–90. Radiother Oncol. 2010;96:48–54. 16. Hardcastle N, Clements N, Chesson B, et al. Results of patient speci- fic quality assurance for patients undergoing stereotactic ablative SUPPORTING INFORMATION radiotherapy for lung lesions. Australas Phys Eng Sci Med. 2014;37:45–52. Additional supporting information may be found online in the 17. Kron T, Chesson B, Hardcastle N, et al. Credentialing of radiotherapy Supporting Information section at the end of the article. centres in Australasia for TROG 09.02 (Chisel), a Phase III clinical trial on stereotactic ablative body radiotherapy of early stage lung cancer. Br J Radiol. 2018;91:20170737.

Journal

Journal of Applied Clinical Medical PhysicsPubmed Central

Published: Feb 13, 2020

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create folders to
organize your research

Export folders, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off