- Email: [email protected]
Dosimetric evaluation of 4 different treatment modalities for curative-intent stereotactic body radiation therapy for isolated thoracic spinal metastases
Medical Dosimetry ] (2016) ]]]–]]]
Medical Dosimetry journal homepage: www.meddos.org
Dosimetric evaluation of 4 different treatment modalities for curative-intent stereotactic body radiation therapy for isolated thoracic spinal metastases Jun Yang, M.D.,*† Lin Ma, M.D., Ph.D.,*‡ Xiao-Shen Wang, M.D.,* Wei Xu Xu, M.D.,* Xiao-Hu Cong, M.D.,* Shou-Ping Xu, Ph.D.,* Zhong-Jian Ju, M.D.,* Lei Du, M.D.,‡ Bo-Ning Cai, M.D.,* and Jack Yang, Ph.D.§ Department of Radiation Oncology, Chinese PLA General Hospital, 28 Fuxing Road, Beijing, 100853, China; †Department of Oncology, First Affiliated Hospital of Xinxiang Medical University, 88 Jiankang Road, Weihui, Henan, 453100, China; ‡Department of Radiation Oncology, Hainan Branch of Chinese PLA General Hospital, Haitang Bay, Sanya, 572000, China; and §Department of Radiation Oncology, Monmouth Medical Center, 300 2nd Avenue, Long Branch, NJ 07740, USA
*
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 May 2015 Received in revised form 28 September 2015 Accepted 14 October 2015
To investigate the dosimetric characteristics of 4 SBRT-capable dose delivery systems, CyberKnife (CK), Helical TomoTherapy (HT), Volumetric Modulated Arc Therapy (VMAT) by Varian RapidArc (RA), and segmental step-and-shoot intensity-modulated radiation therapy (IMRT) by Elekta, on isolated thoracic spinal lesions. CK, HT, RA, and IMRT planning were performed simultaneously for 10 randomly selected patients with 6 body types and 6 body þ pedicle types with isolated thoracic lesions. The prescription was set with curative intent and dose of either 33 Gy in 3 fractions (3F) or 40 Gy in 5F to cover at least 90% of the planning target volume (PTV), correspondingly. Different dosimetric indices, beam-on time, and monitor units (MUs) were evaluated to compare the advantages/disadvantages of each delivery modality. In ensuring the dose-volume constraints for cord and esophagus of the premise, CK, HT, and RA all achieved a sharp conformity index (CI) and a small penumbra volume compared to IMRT. RA achieved a CI comparable to those from CK, HT, and IMRT. CK had a heterogeneous dose distribution in the target as its radiosurgical nature with less dose uniformity inside the target. CK had the longest beam-on time and the largest MUs, followed by HT and RA. IMRT presented the shortest beam-on time and the least MUs delivery. For the body-type lesions, CK, HT, and RA satisfied the target coverage criterion in 6 cases, but the criterion was satisfied in only 3 (50%) cases with the IMRT technique. For the body þ pedicle-type lesions, HT satisfied the criterion of the target coverage of Z90% in 4 of the 6 cases, and reached a target coverage of 89.0% in another case. However, the criterion of the target coverage of Z90% was reached in 2 cases by CK and RA, and only in 1 case by IMRT. For curative-intent SBRT of isolated thoracic spinal lesions, RA is the first choice for the body-type lesions owing to its delivery efficiency (time); the second choice is CK or HT; HT is the preferential choice for the body þ pedicle-type lesions. This study suggests further clinical investigations with longer follow-up for these studied cases. & 2016 American Association of Medical Dosimetrists.
Keywords: Stereotactic body radiation therapy Spinal metastases IMRT Dosimetry
Introduction According to the cancer registry, when late-stage patients presented for radiation therapy, approximately 5% to 10% of patients with cancer developed spinal lesions, and among them
Reprint requests to Jack Yang, Ph.D., Radiation Oncology, Monmouth Medical Center, 300 2nd Ave, Long Branch, NJ. E-mail: http://dx.doi.org/10.1016/j.meddos.2015.10.003 0958-3947/Copyright Ó 2016 American Association of Medical Dosimetrists
thoracic spinal lesions account for approximately 60% to 70% cases.1,2 Radiation therapy has long been established as an effective treatment modality for spinal tumors.3-5 With the advancement of image-guided radiation therapy technique, LINAC-based hypofractionated extracranial radiosurgery or stereotactic body radiation therapy (SBRT) was developed in the mid1990s as an effective and safe treatment modality for spinal tumors, for both primary and metastatic lesions.6,7 Recently, interest in treating solitary and oligo lesions has been increasing
2
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
via various SBRT approaches. It has been shown that the progress of the tumor in the metastatic region could be prevented or delayed by curative-intent SBRT technique. This is also known to improve the survival rate ultimately through improved local control in selected patients.8 Many new LINAC-based treatment machines, with different planning and treatment techniques for SBRT, have been rapidly developing in recent years. The conventional LINAC with add-on modules (Varian and Elekta systems), or newly developed systems with various platforms such as TomoTherapy, CyberKnife (CK), Vero, and Viewray have been engaged into the clinical SBRT programs and services. We have organized and summarized this study to compare the dosimetric characteristics of 4 routinely used treatment modalities in our center—robotic stereotactic radiosurgery via CK, Helical TomoTherapy (HT), volumetric modulated arc therapy in RadipArc (RA) by a Varian system, and segmental intensity-modulated radiation therapy (IMRT) with an Elekta Synergy system, in the SBRT of isolated thoracic lesions planning evaluations. Methods and Materials Delivery systems engaged in study G4 CK system The G4 CK system consists of a robotic manipulator that moves along 6 spatial axes, a target-localization system, and a 6-MV X-band LINAC.9 The performance of the target-locating system is evaluated for mechanical accuracy using both cameras’ alignment and quality assurance tests of the x-ray generators with the flat-panel detectors. The treatment delivery is performed by nonisocentric mode, representing treatment geometry similar to those achieved using multiple pencil beams with 12 fixed-type collimators ranging from 5 to 60 mm. A modulatedintensity fluence pattern from planning could be delivered using multiple beams directed toward unique points within the planning target volume (PTV), each of which has an independent field size and beam weight. This is similar to the field in field techniques which were quite often adopted during the breast treatment to generate uniform dosimetry throughout the whole breast.10 TomoTherapy system The TomoTherapy Hi-Art system (HT) is a special-form IMRT delivery technique.10-12 HT is a rotating radiation treatment machine with a 6-MV linear accelerator mounted on rotating computed tomography (CT) gantry with binary multileaf collimators (MLCs). The 64-binary MLCs with a leaf width of 6.25 mm projected at the isocenter, which was 85 cm away from the x-ray photon source. The TomoTherapy unit in our department used exclusively IMRT procedures. Megavoltage CT (MVCT) images were taken before each treatment to verify the patient’s anatomy with lower beam output in a helical delivery beam. The end result could be reduced artifacts for high-z material such as prostheses, with better soft tissue contrast due to the elimination of photoelectric effect for photon interaction. The couch speed defined for fine imaging was approximately 4 mm per rotation compared with that of normal mode, which was 8 mm per rotation by definition. The coarse speed was defined at 12 mm per rotation, which could degrade the MVCT image quality. Each slice was reconstructed with half-gantry rotation, and the distance between reconstructed slices was 2 mm for fine mode, 4 mm for normal mode, and 6 mm for coarse mode. Acquired images were either manually or automatically registered with fusion planning CT data sets to determine the relative shifts for the patient repositioning before any actual treatment delivery. After performing the image verification, patient position was again verified with correct couch coordinates, and moved to the desired treatment location for treatment delivery. This system is equipped with xenon detectors similar to a CT scanner designed to obtain MVCT images for the image-guided radiation therapy usage of pretreatment setup verification and adaptive therapy. Our HT was installed in 2007; this is the first system in China and we have accumulated considerable clinical cases for comparison purpose. RapidArc system To compete with HT and other clinical advanced systems, the RA system has been studied and implemented by Varian Medical Systems to perform the VMAT treatment, which delivers treatment fractions possessing the characteristics of delivering single or multiple arcs.13 The technical advancement consisted of using varying dose rate, changing gantry rotation speed, and bidirectional MLC motion with variable MLC speed. This system consists of 40 leaf pairs at the center, each with a leaf width of 5 mm, and 20 leaf pairs on each side with a leaf width of 10 mm; therefore, the total number of leaves is 120 for a maximum field size of 40 40 cm2 treatment field. Volumetric image guidance is realized with kV cone-
beam CT guidance. The most significant differences between conventional IMRT and VMAT are the monitor units (MUs) required for delivering the prescribed dose per fraction and beam-on times from early study.14
Synergy S system The Synergy S system consists of a LINAC with step-and-shoot IMRT function and a high-resolution MLC with 40 leaf pairs with a leaf width of 4 mm (Beam modulator, Elekta, Crawley, UK).15 The S system is designed specifically for treating small field size targets with fine leaf resolution. Volumetric image guidance is performed with the cone-beam CT technique (Elekta XVI, Crawley, UK) and the standard segmental IMRT delivery. This platform has been implemented with the traditional segmental IMRT treatment at our facility since early 2000s.
Patients simulation and planning In all, 10 randomly selected patients with 12 isolated targets were under investigation. The involved vertebras in a target volume (TV) had only single-level, class I thoracic vertebral lesions, according to the spinal malignancy classification by DeWald et al.16 These patients were treated with CK using the Xsight Spine tracking system in our center as the baseline modality. Informed consents were obtained from all patients before receiving treatments. Among the selected lesions, 6 of the 12 lesions involved only vertebral body (body type) and 6 others involved vertebral body and unilateral pedicle (body þ pedicle type). Patients’ physical characteristics are summarized in Table 1. Key factors were patients’ KARNOFSKY Performance Status Z 70, life expectancy Z6 months, without symptomatic cord compression, vertebral compression fracture, epidural involvement, or prior spine irradiation. The minimum distance between the gross tumor and spinal cord (on magnetic resonance [MR] images) was Z2 mm. Before CK planning started, contrast-enhanced planning CT scanning (Brilliance CT Big Bore, Philips, Andover, MA) with 1.5-mm slice thickness and contrastenhanced 3 dimensional planning MR imaging (Discovery 1.5 T, GE, Milwaukee, WI) with FSPGR sequence of 1.2-mm slice thickness were performed in the supine treatment position for each patient. Planning CT and MR images were fused for gross tumor volume (GTV) and cord contouring. MR imaging with plain T2 sequence was also used for target definition. A treatment plan was generated using the CK MultiPlan 4.0.2 nonisocentric inverse-planning algorithm with tissue density heterogeneity corrections. As the CK beam was not flattened, 2 sets of collimators were used to maintain homogeneity. Typical beam and time reduction techniques were also implemented toward the end of the optimization process to produce a practically deliverable plan. Patients in the supine position were then treated by fitting into a vacuum cushion, taking advantages of fiducial-free spine tracking methodology. SBRT cases were also replanned with a HT 4.0.4 planning workstation for the HT system, with an Eclipse v10 workstation (Varian Medical Systems) for RA, and with Pinnacle3 8.0 (Philips Medical System) for 9 fields segmental IMRT. For the finest calculations, HT plans employed 1.0-cm jaw and 0.43-pitch combination to simulate the comparable plans from CK. Also RA plans employed dual arcs, from 179.91 to 180.11 counterclockwise and from 180.11 to 179.91 in the reverse direction to form 2 complete arc sweeps. The GTV was defined as the metastatic lesion within a vertebral body (body type) or within a vertebral body and its unilateral pedicle (body þ pedicle type). A safety margin expansion of 3 mm from GTV was contoured as the PTV. PTV was limited by the cord if overlapped, which was defined as the cord contoured from planning MR images plus an expansion of 2 mm. Cord and esophagus were extended 6 mm above and below PTV in craniocaudal direction, for accrual dosevolume histograms analysis.
Dosimetric evaluations Total treatment doses prescribed were 33 Gy/3F and 40 Gy/5F to PTV for bodytype and pedicle-type lesions, with a biologic effective dose (BED10) (α/β ¼ 10 Gy) of 69.3 Gy and 72 Gy, respectively. For target coverage (VPTV), 100% of the prescription dose line has to cover Z90% of the PTV. The dose-volume constraints for cord and esophagus also followed the protocol documented by Timmerman.17 Patients were then treated on consecutive days, with 1 fraction (F) per day until completion. The dose-volume histograms provided necessary information on dose distribution for comparing treatment plans with each modality. Based on these calculated histograms, gradient index (GI), dose heterogeneity index (HI), conformity index (CI), and beam-on time have been compiled for evaluation and comparison. The GI has been introduced to reflect the dose falloff outside the target. The low dose outside the prescription isodose volume (PIV) may cover significant amounts of normal tissues and can be responsible for complications, especially when the target is near critical structures. This index is defined as the ratio of the volume of half the prescription dose to the PIV.18 A low GI indicates a low dose spread outside the lesion and a rapid dose falloff. The Paddick’s CI factor,19 which takes into account the location of the PIV with respect to the TV, is defined as follows: CI ¼ (TV covered by PIV)2/(TV PIV). This
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
3
Table 1 Patients' clinical and dosimetric characteristics Patient Sex Age (y)
Primary tumor
Metastatic site
1 2 3
F M M
43 85 75
4 5 6-1
M M M
56 48 84
6-2
M
84
7-1
M
73
NSLC T10 NSLC T5 Renal adeno T7 carcinoma NSLC T9 NSLC T10 Prostate T7 adenocarcinoma Prostate T9 adenocarcinoma NSLC T3
7-2
M
73
NSLC
T5
8
M
65
T9
9
M
54
10
F
75
Rectal adenocarcinoma Stomach adenocarcinoma NSLC
T5 T5
Metastatic type
PTV volume (mm3)
VPTV (%) (CK)
VPTV (%) (HT)
VPTV (%) (RA)
VPTV (%) (IMRT)
Shortest distance PTV-cord (mm)
Shortest distance PTV-esophagus (mm)
Body Body Body
7427.22 23,483.56 23,956.25
95.2 94.2 90.5
95.1 91.7 92.1
96.2 91.9 93.5
91.7 87.8 88.7
1.18 0.62 0
9.1 3.3 2.0
Body Body Body
27,317.05 17,425.95 27,055.38
90.5 93.9 93.9
93.6 95.6 99.3
94.4 95.5 96.9
87.6 93.0 96.1
0 0 2.98
7.7 20 3.7
Body þ pedicle Body þ pedicle Body þ pedicle Body þ pedicle Body þ pedicle Body þ pedicle
19,990.05
95.5
94.9
98.6
90.7
3.24
12.6
4582.00
91.1
90.3
88.8
87.5
0
2.4
13,358.93
81.2
89.0
91.9
82.3
0
4.0
28,616.49
89.8
91.2
87.2
88.8
0
23.2
22,743.23
74.6
84.2
85.1
78.3
0
0
30,918.72
79.8
90.0
87.3
87.2
0
7.8
NSLC = non-small cell lung carcinoma. VPTV: PTV coverage.
n
index analyzes both the quality of the coverage and the volume of healthy tissue receiving a dose larger than or equal to that in the PIV. With the CI definition, a perfect plan would yield a CI equal to 1.00 and would be equal to the prescription isodose sculpted around the TV. The HI is defined as the ratio of the difference between the maximum and minimum dose to the mean dose of the PTV. This means that when the index is 0, the dose distribution to the PTV would be the most homogeneous, and increasing HI indicates increasing dose heterogeneity. Definition of V20% to 80% reflects the penumbra volume. The maximum dose (Dmax), minimum dose (Dmin), mean dose (Dmean), D50 (the dose received by 50% of the PTV volume), D98 (the dose received by 98% of the PTV volume) of the PTV, esophageal D5 (the dose received by 5% of the esophageal volume), cord D1 (the dose received by 1% of the cord volume), and total beam-on time of each dedicated treatment system were all reported and compared. The delivery dose rates of CK, HT, RA, and IMRT were 800, 900, 600, and 400 MU/min, respectively. Statistical analyses Various statistical analysis tools, such as 1-way analysis of variance test, KruskalWallis test, Bonferroni test, and LSD-t test (Statistical Package for the Social Sciences; IBM Corporation, NY) were implemented to evaluate the statistical significance of the differences between the different indices with all the systems. Therefore, p o 0.05 was considered to indicate statistical significance in our study cases.
Results Patients’ clinical and target characteristics and PTV coverage are presented in Table 1. Tables 2 and 3 summarize the dosimetric indices for body and body þ pedicle types of thoracic spinal lesions and each treatment modality. For the body-type lesions with a prescription dose of 33 Gy/3F to PTV, CK, HT, and RA satisfied VPTV Z 90% in all the 6 cases; however, IMRT reached this criterion in only 3 out of the 6 cases (50%). CK, HT, and RA achieved a sharp GI and a small penumbra volume compared to IMRT (p ¼ 0.001, 0.027, 0.01; p ¼ 0.002, 0.015, 0.001; respectively), without intergroup differences among CK, HT, and RA. RA achieved a comparable CI related to CK, HT, and IMRT (p ¼ 0.001, 0.047, 0.035, respectively), without intergroup differences among CK, HT, and IMRT. CK had a high Dmax, a high Dmean, and a high D50 in the PTV compared with HT, RA, and IMRT (p ¼ 0.000, 0.014, 0.015; p ¼ 0.001, 0.007, 0.002; p ¼ 0.010, 0.000, 0.000; respectively), reflecting a heterogeneous dose distribution in CK planning. The intergroup differences were significant (all with a p o 0.001) between any 2 of the 4 modalities for beam-on time and MU except between RA and IMRT, showing that CK had
the longest beam-on time and the largest MU, followed by HT, RA, and IMRT, which had the shortest beam-on time and the least amount of delivery time. For the body þ pedicle-type metastasis with a prescription dose of 40 Gy/5F to PTV, HT satisfied VPTV Z 90% in 4 of the 6 cases (66.7%), and reached a target coverage of 89.0% in another case. CK and RA reached VPTV Z 90% in 2 cases (33.3%), and only in 1 case (16.7%) by IMRT. As the results obtained for the body-type metastasis, CK, HT, and RA achieved a sharp GI and a small penumbra volume compared to IMRT (p ¼ 0.000, 0.000, 0.000; p ¼ 0.021, 0.066, 0.007; respectively), without intergroup differences among CK, HT, and RA. RA achieved a good CI compared with CK, HT, and IMRT (all had a p ¼ 0.000), without intergroup differences among CK, HT, and IMRT. The intergroup differences were significant (p ¼ 0.000 between CK and HT, p ¼ 0.035 between CK and RA, p ¼ 0.017 between CK and IMRT; p ¼ 0.000 between HT and RA, also between HT and IMRT) between any 2 of the 4 modalities for D50 in the PTV except between RA and IMRT, showing that CK had the highest figure of merit, followed by HT, RA, and IMRT, which had the lowest figure of merit. The intergroup differences were significant (all had p o 0.001) between any 2 of the 4 modalities for beam-on time except between HT and IMRT, and between RA and IMRT. The intergroup differences were significant (all had p o 0.001, except p ¼ 0.031 between HT and RA) between any 2 of the 4 modalities for MU except between HT and IMRT, and between RA and IMRT, showing that CK had the longest delivery time, followed by HT, RA, and IMRT, which had the shortest delivery time. A very typical dosimetry outcome can be observed in Fig. 1, and the difference between body-type and body þ pedicle-type TV is shown in Fig. 2.
Discussion From the past, 2-dimensional conventional radiation therapy has been a well-established palliative treatment paradigm for painful vertebral lesions, with a median duration of pain relief of approximately 3 to 6 months after irradiation. Although no difference was detected in pain response rate (from 50% to 90%) and dose-response relationship for pain control between single- and multifraction schedules in prospective studies, single fractionation
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
4
Table 2 Dosimetric comparison (33 Gy/3F to PTV) in body-type thoracic spinal metastases Dosimetric factor
CK
HT
RA
IMRT
PTV GI Mean SD
3.33 0.24
3.80 0.50
3.67 0.30
4.70 0.75
PTV CI Mean SD
0.63 0.08
0.79 0.14
0.98 0.22
0.77 0.17
PTV HI Mean SD
0.73 0.08
0.57 0.22
0.57 0.21
0.64 0.12
427.25 155.42
522.77 169.64
395.04 139.84
789.54 216.98
PTV Dmax (Gy) Mean SD
46.16 2.41
39.21 2.87
42.03 3.54
42.06 1.35
PTV Dmin (Gy) Mean SD
17.27 3.93
18.87 6.77
20.60 6.33
18.73 4.20
PTV Dmean (Gy) Mean SD
39.92 2.44
35.64 1.37
37.00 1.68
36.46 0.91
PTV D98 (Gy) Mean SD
25.10 5.66
28.21 3.21
30.20 2.52
26.58 4.03
PTV D50 (Gy) Mean SD
41.44 1.97
36.01 1.60
37.35 1.79
37.03 0.84
Beam-on time (min) Mean SD
55.33 6.95
29.17* 1.47
10.50 3.94
12.83 3.71
43,722.24 4048.97
24,993.50 1359.79
5785.00 1037.72
5077.43 1512.31
22.03 1.38
20.63 2.01
21.27 0.76
22.25 2.50
17.92 3.68
15.97 3.73
17.85 1.39
17.55 2.48
Esophagus Dmax (Gy) Mean SD
24.87 1.69
23.48 4.07
22.60 4.00
24.59 3.49
Esophagus D5 (Gy) Mean SD
18.75 1.32
17.69 3.25
17.43 3.44
15.56 6.02
PTV V20% Mean SD
to 80%
(mm3)
MU Mean SD Cord Dmax (Gy) Mean SD Cord D1 (Gy) Mean SD
F
p 12.322
0.006
4.925
0.010
1.133
0.360
6.432
0.003
6.916
0.002
0.375
0.772
7.239
0.002
1.771
0.185
13.295
0.000
128.523
0.000
374.833
0.000
1.026
0.402
0.568
0.643
0.460
0.714
0.429
0.934
SD ¼ standard deviation. n
In 2 times.
has been associated with a 20% to 25% incidence of repeat treatment vs 7% to 8% with fractionated radiation therapy. Patients treated with lower biologic effective dose were more frequently given second courses of radiation to the metastatic sites.20-23 Longcourse and higher dose radiation therapy for metastatic spinal
cord compression was significant for improved progression-free survival and local control.24 With the emerging targeted systemic therapies and much better understanding of cancer biology, patients might live longer with bony lesions. There is a good rationale for high dose escalation beyond palliative dose levels to
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
5
Table 3 Dosimetric comparisons (40 Gy/5F to PTV) for body þ pedicle-type thoracic spinal metastases Dosimetric factor
CK
HT
RA
IMRT
PTV GI Mean SD
3.23 0.22
3.58 0.43
3.95 0.35
6.33 1.32
PTV CI Mean SD
0.54 0.09
0.58 0.15
0.92 0.10
0.60 0.14
0.86 0.19
0.73 0.28
0.74 0.27
0.94 0.18
407.68 214.20
474.47 176.22
347.80 155.22
705.96 262.95
57.81 2.93
54.25 9.87
57.33 6.22
57.37 5.81
15.76 7.65
20.04 6.86
23.00 8.41
15.00 5.45
PTV Dmean (Gy) Mean SD
49.16 2.68
45.88 3.99
45.86 1.52
45.22 2.99
PTV D98 (Gy) Mean SD
26.60 6.68
29.65 5.44
31.97 5.29
25.80 5.91
PTV D50 (Gy) Mean SD
52.28 2.09
47.15 4.59
46.15 1.40
46.35 2.55
Beam-on time (min) Mean SD
61.67 8.14
20.17* 2.86
6.30 0.52
13.67 3.39
61,387.27 14527.16
17,258.00 2653.29
3807.00 326.10
5524.62 1373.21
Cord Dmax (Gy) Mean SD
31.00 4.27
28.49 1.55
28.78 1.36
27.12 1.88
Cord D1 (Gy) Mean SD
22.25 4.06
20.64 2.95
23.71 2.46
18.48 2.89
Esophagus Dmax (Gy) Mean SD
28.60 6.01
24.14 8.24
27.71 7.11
29.75 6.55
Esophagus D5 (Gy) Mean SD
22.95 6.24
19.47 7.59
21.73 5.84
23.62 8.16
PTV HI Mean SD V20% to 80% (mm3) Mean SD PTV Dmax (Gy) Mean SD PTV Dmin (Gy) Mean SD
MU Mean SD
n
F
p
17.257
0.001
11.941
0.000
1.095
0.374
3.466
0.036
0.363
0.781
1.643
0.211
2.219
0.117
1.416
0.268
37.664
0.000
21.161
0.000
21.147
0.000
5.340
0.149
3.056
0.052
0.596
0.627
0.338
0.798
In 2 times.
improve on existing rates of local and pain control, especially for those patients with solitary or oligo lesions and with a relatively indolent or well-controlled primary tumor, good performance status, as well as absence of visceral lesions. Recent advances in radiotherapy technology—stereotactic radiosurgery in generating sharp dose gradient, in combination with IMRT for highly conformal dose distributions and image tracking and guidance for
precise treatment delivery, have made curative-intent SBRT of spinal lesions possible and early results of pain and local tumor control were promising.25,26 The prospective study of Milano et al.8 concluded that bony oligo lesions and solitary metastasis from breast cancer were well controlled by the SBRT technique, with a fourfold and threefold reduced hazard of death, respectively. With our clinical planning evaluation of SBRT, to our best knowledge,
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
6
A
B
C
D
Fig. 1. Axial view of SBRT (40 Gy/5F) plan of a body þ pedicle-type metastasis (T9). PTV is contoured in red, cord in sky blue, and esophagus in white. (A) CyberKnife (VPTV ¼ 89.8%). (B) Helical tomotherapy (VPTV ¼91.2%). (C) RapidArc (VPTV ¼87.2%). (D) IMRT (VPTV ¼88.8%). Isodose lines: 50 Gy in black, 40 Gy in green, 30 Gy in yellow, 20 Gy in blue, and 10 Gy in pink. (Color version of figure is available online.)
this might be the first comparative dosimetric study on 4 different LINAC-based treatment modalities for isolated thoracic spinal metastasis SBRT.
For a curative-intent SBRT, high dose and conformal coverage of TV are critical. Yamada et al.27 suggested that radiation dose was a significant predictor of local control after SBRT for metastatic
A
B
C
D
E
F
Fig. 2. Illustration in 3 dimensions of a body-type metastasis (A-C) and a body þ pedicle-type metastasis (D-F). A margin of 3 mm from GTV (in green) is defined as PTV (in red). (Color version of figure is available online.)
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
spinal lesions. Koyfman et al.28 treated 208 spinal metastatic lesions on 149 patients with 10 to 16 Gy in a single fraction, and found that the incidence of marginal recurrence increased with a dose of o16 Gy. Choi et al.29 had treated 20 patients with gastrointestinal tract cancer with HT for 31 spinal oligo lesions, which showed that BED10 higher than 57 Gy to GTV could improve local control. However, an adverse event after spinal SBRT was vertebral compression fracture, whose risk increased significantly with a dose per fraction of 20 Gy or greater.30,31 Therefore, to summarize all the available clinical data, we chose to prescribe 33 Gy in 3F and 40 Gy in 5F to cover at least 90% of PTV volume, for body and body þ pedicle type, respectively. These prescriptions could translate the BED10 to 69.3 Gy and 72 Gy, correspondingly. In the process of treatment planning, as organs at risk (OARs) (e.g., cord and esophagus in this study) were protected with first priority, all the dosimetric indices for OARs showed no significant differences among the 4 modalities for sparing purposes in planning results. For the body-type lesions, CK, HT, and RA reached PTV coverage criterion in all the 6 cases; however, IMRT reached this criterion only in 50% of the cases. CK, HT, and RA also achieved a sharp GI and a small penumbra volume compared to IMRT, without intergroup differences among these 3 modalities. The clinical choice would be focused on the first 3 modalities. As RA achieved a good CI and at the same time had the shortest beam-on time and least MU compared with CK and HT, it would be the first choice for the body-type isolated spinal lesions. The second choice would be CK or HT. Although HT had a homogeneous dose distribution and shorter beam-on time compared with CK, a fraction of high-dose HT SBRT would be divided into 2 separated subfractions by the fractional dose constraint of 6 Gy in the HT system, which has no intrafraction target-tracking system like CK, so intrafractional movements and errors could take place during interfaction. To resolve this clinically relevant issue, one could consider a secondary MVCT image guidance before the second subfraction, which takes 3 to 5 minutes more. For the body þ pedicle-type lesions that had an irregular and large PTV volume compared with the body-type metastasis, HT completely satisfied target coverage in 4 of the 6 cases (66.7%) and reached 89.0% coverage of the PTV in another patient; CK and RA reached 90% coverage criterion in 2 cases (33.3%), and only in 1 case by IMRT. Like the case of body-type metastasis, CK, HT, and RA all achieved a sharp GI and a small penumbra volume compared to IMRT, without intergroup differences among the first 3 modalities. HT would be the preferential choice for the body þ pedicle-type isolated spinal metastasis, but at the cost of treatment time and MU compared with the RA technique. CK is a cone-based LINAC system that covered the lesion with nonisocentric noncoplanar circular beams spread over a large range of angles. With the use of multiple entries of different sizes, modulated beams, and the employment of inverse planning based on specific constraints, CK could achieve comparable conformity, while reducing significantly the treatment time compared with a gamma-knife system for benign intracranial lesions.32 Unlike the pretreatment image guidance of HT, RA, and Synergy S systems, CK is unique in that it uses intrafractional X-ray imaging (typically every 30 to 60 seconds), and automatic LINAC positional adjustments to compensate for any detected changes in target positioning. For relatively regular-shaped body-type metastasis, CK achieved high target coverage, sharp GI, and small penumbra volume like HT and RA, with a heterogeneous dose distribution, which however might not be a disadvantage as the hot spots were located within the TV. HT is a modality for delivering IMRT treatments using a rotating LINAC mounted on a continuously moving slip ring gantry in synchrony with the couch motion. This technique delivers highly conformal dose using 6-MV x-ray beams with a 64 leaves binary collimator of a 40-cm-
7
wide fan of thicknesses 1.0 to 5.0 cm to an isocenter 85 cm away from the source. The MLC leaves open 51 times per rotation as the gantry moves at a constant speed.10-12 In this dosimetric study of isolated thoracic spinal metastasis SBRT, with the high modulation resolution of its 1.0-cm jaw thickness, HT achieved a high target coverage even in complex-shaped body þ pedical-type lesions. In this study, the VMAT technique was accomplished by the RA system, which used more beam directions than fixed-gantry IMRT did and delivered highly conformal volumetric dose distributions in dual 3601 arcs. In addition to the marked advantage of its shorter treatment time and fewer MUs compared with other modalities of LINAC-based therapy, the combination of the advantages of arc therapy and IMRT, as demonstrated by Chin et al.33 in treatment planning with collimator motion, dose-rate variation, and various arc numbers, the highly conformal dose distributions could be achieved also for various target shapes, making RA a feasible choice for the body-type metastasis. In our results, RA could achieve relatively high target coverage, sharp GI, and small penumbra volume comparable to those of CK and HT, and also with an acceptable CI. However, the extra imaging time should be considered in comparing with similar technologies. The fine-leaf-width MLC in combination with IMRT technique could also yield dosimetric benefits in SBRT. Treatment of small lesions in cases involving complex target and OARs geometry would benefit from use of a fine-leaf-width MLC. In a previous dosimetric study for IMRT of spinal tumors, Wu et al.34 showed that the 2.5-mm leaf-width collimator significantly improved spinal cord sparing, with dose reductions of 14% to 19% in highto mid-level dose regions, compared with the 5-mm leaf width collimator. Chae et al.35 showed that smaller MLC leaf width provided improved target coverage in both IMRT and VMAT for spinal lesions, and its improvement was larger in IMRT than in VMAT. In addition, the smaller MLC leaf width was more effective for conforming to the complex-shaped targets. In this study, stepand-shoot IMRT-based SBRT was performed with the Synergy S system, which has a 4-mm leaf-width MLC. For small lesions, the limitation and the step-and-shoot beam path may not be efficient for spinal SBRT with our analysis, especially for the complexshaped body þ pedicle-type lesions, whose dose gradient requirement is a critical factor. In this case the current Synergy S system might not be able to meet the strict dosimetric requirements.
Conclusions All modalities are capable of performing SBRT based on the comparable GI and conformal index. Based on this dosimetric study of different LINAC-based SBRT modalities with isolated thoracic spinal metastasis, treatment delivery technique based on our planning results could be properly recommended: the RA is the first choice for the body-type metastasis owing to its efficiency (shortest treatment time) and similar dosimetric outcomes compared with other modalities; the second choice is CK or HT. However, HT is the preferential choice for the body þ pedicletype metastasis from its planning results. This study suggests further clinical investigation with clinical outcomes and systematic evaluations. As the results indicated, the step-and-shoot IMRT was clinically limited by its beam flexibility and dose gradient might not fulfill the dosimetric constraints in our study. Depending on the available technology, this conclusion was applied only to our center with current delivery systems addressed in the paper. Authors’ Contributions Ju.Y. participated in acquisition of data, performed the statistical analyses, and helped write the manuscript. L.M. and Ju.Y. were
8
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
responsible for study conception and design, and drafted the manuscript. X.S.W., W.X., and X.H.C. contributed to data analyses. S.P.X. and Z.J.J. participated in the design and coordination of the study. L.D. and B.N.C. participated in acquisition of data. Ja.Y. reviewed and modified the contents and references. All authors read and approved the final manuscript. References 1. Gabriel, K.; Schiff, D. Metastatic spinal cord compression by solid tumors. Semin. Neurol. 24:375–83; 2004. 2. Schiff, D. Spinal cord compression. Neurol. Clin. 21:67–86; 2003. 3. Ryu, S.I.; Chang, S.D.; Kim, D.H.; et al. Image-guided hypo-fractionated stereotactic radiosurgery to spinal lesions. Neurosurgery 49:838–46; 2001. 4. Faul, C.M.; Flickinger, J.C. The use of radiation in the management of spinal lesions. J. Neurooncol. 23:149–61; 1995. 5. Sundaresan, N.; Digiacinto, G.V.; Hughes, J.E.O.; et al. Treatment of neoplastic spinal cord compression: Results of a prospective study. Neurosurgery 29:645–50; 1991. 6. Blomgren, H.; Lax, I.; Naslund, I.; et al. Stereotactic high dose fraction radiation therapy of extracranial tumors using an accelerator. Clinical experience of the first thirty-one patients. Acta Oncol. 34:861–70; 1995. 7. Hamilton, A.J.; Lulu, B.A.; Fosmire, H.; et al. Preliminary clinical experience with linear accelerator-based spinal stereotactic radiosurgery. Neurosurgery 36:311–9; 1995. 8. Milano, M.T.; Katz, A.W.; Zhang, H.; et al. Oligometastases treated with stereotactic body radiotherapy: long-term follow-up of prospective study. Int. J. Radiat. Oncol. Biol. Phys. 83:878–86; 2012. 9. Antypas, C.; Pantelis, E. Performance evaluation of a CyberKnife G4 imageguided robotic stereotactic radiosurgery system. Phys. Med. Biol. 53:4697–718; 2008. 10. Kestin, L.L.; Sharpe, M.B.; Frazier, R.C.; et al. Intensity modulation to improve dose uniformity with tangential breast radiotherapy: Initial clinical experience. Int. J. Radiat. Oncol. Biol. Phys. 48(5):1559–68; 2000. 11. Mackie, T.R.; Holmes, T.; Swerdloff, S.; et al. TomoTherapy: A new concept for the delivery of dynamic conformal radiotherapy. Med. Phys. 20:1709–19; 1993. 12. Jeraj, R.; Mackie, T.R.; Balog, J.; et al. Radiation characteristics of helical TomoTherapy. Med. Phys. 31:396–404; 2004. 13. Otto, K. Volumetric modulated arc therapy. IMRT in a single gantry arc. Med. Phys. 35:310–7; 2008. 14. Zhang, P.; Happersett, L.; Hunt, M.; et al. Volumetric modulated arc therapy: Planning and evaluation for prostate cancer cases. Int. J. Radiat. Oncol. Biol. Phys. 76(5):1456–62; 2010. 15. Guckenberger, M.; Sweeney, R.A.; Flickinger, J.C.; et al. Clinical practice of image-guided spine radiosurgery—Results from an international research consortium. Radiat. Oncol. 6:172–82; 2011. 16. Dewald, R.L.; Bridwell, K.H.; Prodromas, C.; et al. Reconstructive spinal surgery as palliation for metastatic malignancies of the spine. Spine 10:21–6; 1985. 17. Timmerman, R.D. An overview of hypofractionation and introduction to this issue of seminars in radiation oncology. Semin. Radiat. Oncol. 18:215–22; 2008.
18. Paddick, I.; Lippitz, B. A simple dose gradient measurement tool to complement the conformity index. J. Neurosurg. 105:194–201; 2006. 19. Paddick, I. A simple scoring ratio to index the conformity of radiosurgical treatment plans. J. Neurosurg. 93:219–22; 2000. 20. Chow, E.; Zeng, L.; Salvo, N.; et al. Update on the systemic review of palliative radiotherapy trials for bone metastasis. Clin. Oncol. (R. Coll. Radiol.) 24:112–24; 2012. 21. Lutz, S.; Berk, L.; Chang, E.; et al. American Society for Radiation Oncology (ASTRO). Palliative radiotherapy for bone lesions: An ASTRO evidence-based guideline. Int. J. Radiat. Oncol. Biol. Phys. 79:965–76; 2011. 22. Wu, J.S.; Wong, R.; Johnston, M.; et al. Meta-analysis of dose-fractionation radiotherapy trials for the palliation of painful bone metastases. Int. J. Radiat. Oncol. Biol. Phys. 55:594–605; 2003. 23. Chow, E.; Harris, K.; Fan, G.; et al. Palliative radiotherapy trials for bone lesions: A systematic review. J. Clin. Oncol. 25:1423–36; 2007. 24. Rades, D.; Lange, M.; Veninga, T.; et al. Preliminary results of spinal cord compression recurrence evaluation (score-1) study comparing short-course versus long-course radiotherapy for local control of malignant epidural spinal cord compression. Int. J. Radiat. Oncol. Biol. Phys. 73:228–34; 2009. 25. Joaquim, A.F.; Ghizoni, E.; Tedeschi, H.; et al. Stereotactic radiosurgery for spinal lesions: A literature review. Einstein (Sao Paulo) 11:247–55; 2013. 26. Sahgal, A.; Larson, D.A.; Chang, E.L. Stereotactic body radiosurgery for spinal lesions: A critical review. Int. J. Radiat. Oncol. Biol. Phys. 71:652–65; 2008. 27. Yamada, Y.; Bilsky, M.H.; Lovelock, D.M.; et al. High-dose, single fraction imageguided intensity-modulated radiotherapy for for metastatic spinal lesions. Int. J. Radiat. Oncol. Biol. Phys. 71:484–90; 2008. 28. Koyfman, S.A.; Djemil, T.; Burdick, M.J.; et al. Marginal recurrence requiring salvage radiotherapy after stereotactic body radiotherapy for spinal lesions. Int. J. Radiat. Oncol. Biol. Phys. 83:297–302; 2012. 29. Choi, Y.L.; Kim, J.W.; Lee, I.J.; et al. Helical TomoTherapy for spine oligometastases from gastrointestinal malignancies. Radiat. Oncol. J. 29:219–27; 2011. 30. Sahgal, A.L.; Atenafu, E.G.; Chao, S.; et al. Vertebral compression fracture after spine stereotactic body radiotherapy: A multi-institutional analysis with a focus on radiation dose and the spinal instability neoplastic score. J. Clin. Oncol. 31:3426–31; 2013. 31. Cunha, M.V.; Al-Omair, A.; Atenafu, E.G.; et al. Vertebral compression fracture (VCF) after spine stereotactic body radiation therapy (SBRT): Analysis of predictive factors. Int. J. Radiat. Oncol. Biol. Phys. 84:e343–9; 2012. 32. Gevaert, T.; Levivier, M.; Lacornerie, T.; et al. Dosimetric comparison of different treatment modalities for stereotactic radiosurgery of arteriovenous malformations and acoustic neuromas. Radiother. Oncol. 106:192–7; 2013. 33. Chin, L.M.; Kijewski, P.K.; Sevesson, G.K.; et al. Dose optimization with computer-controlled gantry rotation, collimator motion and dose-rate variation. Int. J. Radiat. Oncol. Biol. Phys. 9:723–9; 1983. 34. Wu, Q.,J.; Wang, Z.; Kirkpatrick, J.P.; et al. Impact of collimator leaf width and treatment technique on stereotactic radiosurgery and radiotherapy plans for intra- and extracranial lesions. Radiat. Oncol. 4:3–12; 2009. 35. Chae, S.M.; Lee, G.W.; Son, S.H. The effect of multileaf collimator leaf width on the radiosurgery planning for spine lesion treatment in terms of the modulated techniques and target complexity. Radiat. Oncol. 9:72–81; 2014.
Medical Dosimetry journal homepage: www.meddos.org
Dosimetric evaluation of 4 different treatment modalities for curative-intent stereotactic body radiation therapy for isolated thoracic spinal metastases Jun Yang, M.D.,*† Lin Ma, M.D., Ph.D.,*‡ Xiao-Shen Wang, M.D.,* Wei Xu Xu, M.D.,* Xiao-Hu Cong, M.D.,* Shou-Ping Xu, Ph.D.,* Zhong-Jian Ju, M.D.,* Lei Du, M.D.,‡ Bo-Ning Cai, M.D.,* and Jack Yang, Ph.D.§ Department of Radiation Oncology, Chinese PLA General Hospital, 28 Fuxing Road, Beijing, 100853, China; †Department of Oncology, First Affiliated Hospital of Xinxiang Medical University, 88 Jiankang Road, Weihui, Henan, 453100, China; ‡Department of Radiation Oncology, Hainan Branch of Chinese PLA General Hospital, Haitang Bay, Sanya, 572000, China; and §Department of Radiation Oncology, Monmouth Medical Center, 300 2nd Avenue, Long Branch, NJ 07740, USA
*
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 May 2015 Received in revised form 28 September 2015 Accepted 14 October 2015
To investigate the dosimetric characteristics of 4 SBRT-capable dose delivery systems, CyberKnife (CK), Helical TomoTherapy (HT), Volumetric Modulated Arc Therapy (VMAT) by Varian RapidArc (RA), and segmental step-and-shoot intensity-modulated radiation therapy (IMRT) by Elekta, on isolated thoracic spinal lesions. CK, HT, RA, and IMRT planning were performed simultaneously for 10 randomly selected patients with 6 body types and 6 body þ pedicle types with isolated thoracic lesions. The prescription was set with curative intent and dose of either 33 Gy in 3 fractions (3F) or 40 Gy in 5F to cover at least 90% of the planning target volume (PTV), correspondingly. Different dosimetric indices, beam-on time, and monitor units (MUs) were evaluated to compare the advantages/disadvantages of each delivery modality. In ensuring the dose-volume constraints for cord and esophagus of the premise, CK, HT, and RA all achieved a sharp conformity index (CI) and a small penumbra volume compared to IMRT. RA achieved a CI comparable to those from CK, HT, and IMRT. CK had a heterogeneous dose distribution in the target as its radiosurgical nature with less dose uniformity inside the target. CK had the longest beam-on time and the largest MUs, followed by HT and RA. IMRT presented the shortest beam-on time and the least MUs delivery. For the body-type lesions, CK, HT, and RA satisfied the target coverage criterion in 6 cases, but the criterion was satisfied in only 3 (50%) cases with the IMRT technique. For the body þ pedicle-type lesions, HT satisfied the criterion of the target coverage of Z90% in 4 of the 6 cases, and reached a target coverage of 89.0% in another case. However, the criterion of the target coverage of Z90% was reached in 2 cases by CK and RA, and only in 1 case by IMRT. For curative-intent SBRT of isolated thoracic spinal lesions, RA is the first choice for the body-type lesions owing to its delivery efficiency (time); the second choice is CK or HT; HT is the preferential choice for the body þ pedicle-type lesions. This study suggests further clinical investigations with longer follow-up for these studied cases. & 2016 American Association of Medical Dosimetrists.
Keywords: Stereotactic body radiation therapy Spinal metastases IMRT Dosimetry
Introduction According to the cancer registry, when late-stage patients presented for radiation therapy, approximately 5% to 10% of patients with cancer developed spinal lesions, and among them
Reprint requests to Jack Yang, Ph.D., Radiation Oncology, Monmouth Medical Center, 300 2nd Ave, Long Branch, NJ. E-mail: http://dx.doi.org/10.1016/j.meddos.2015.10.003 0958-3947/Copyright Ó 2016 American Association of Medical Dosimetrists
thoracic spinal lesions account for approximately 60% to 70% cases.1,2 Radiation therapy has long been established as an effective treatment modality for spinal tumors.3-5 With the advancement of image-guided radiation therapy technique, LINAC-based hypofractionated extracranial radiosurgery or stereotactic body radiation therapy (SBRT) was developed in the mid1990s as an effective and safe treatment modality for spinal tumors, for both primary and metastatic lesions.6,7 Recently, interest in treating solitary and oligo lesions has been increasing
2
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
via various SBRT approaches. It has been shown that the progress of the tumor in the metastatic region could be prevented or delayed by curative-intent SBRT technique. This is also known to improve the survival rate ultimately through improved local control in selected patients.8 Many new LINAC-based treatment machines, with different planning and treatment techniques for SBRT, have been rapidly developing in recent years. The conventional LINAC with add-on modules (Varian and Elekta systems), or newly developed systems with various platforms such as TomoTherapy, CyberKnife (CK), Vero, and Viewray have been engaged into the clinical SBRT programs and services. We have organized and summarized this study to compare the dosimetric characteristics of 4 routinely used treatment modalities in our center—robotic stereotactic radiosurgery via CK, Helical TomoTherapy (HT), volumetric modulated arc therapy in RadipArc (RA) by a Varian system, and segmental intensity-modulated radiation therapy (IMRT) with an Elekta Synergy system, in the SBRT of isolated thoracic lesions planning evaluations. Methods and Materials Delivery systems engaged in study G4 CK system The G4 CK system consists of a robotic manipulator that moves along 6 spatial axes, a target-localization system, and a 6-MV X-band LINAC.9 The performance of the target-locating system is evaluated for mechanical accuracy using both cameras’ alignment and quality assurance tests of the x-ray generators with the flat-panel detectors. The treatment delivery is performed by nonisocentric mode, representing treatment geometry similar to those achieved using multiple pencil beams with 12 fixed-type collimators ranging from 5 to 60 mm. A modulatedintensity fluence pattern from planning could be delivered using multiple beams directed toward unique points within the planning target volume (PTV), each of which has an independent field size and beam weight. This is similar to the field in field techniques which were quite often adopted during the breast treatment to generate uniform dosimetry throughout the whole breast.10 TomoTherapy system The TomoTherapy Hi-Art system (HT) is a special-form IMRT delivery technique.10-12 HT is a rotating radiation treatment machine with a 6-MV linear accelerator mounted on rotating computed tomography (CT) gantry with binary multileaf collimators (MLCs). The 64-binary MLCs with a leaf width of 6.25 mm projected at the isocenter, which was 85 cm away from the x-ray photon source. The TomoTherapy unit in our department used exclusively IMRT procedures. Megavoltage CT (MVCT) images were taken before each treatment to verify the patient’s anatomy with lower beam output in a helical delivery beam. The end result could be reduced artifacts for high-z material such as prostheses, with better soft tissue contrast due to the elimination of photoelectric effect for photon interaction. The couch speed defined for fine imaging was approximately 4 mm per rotation compared with that of normal mode, which was 8 mm per rotation by definition. The coarse speed was defined at 12 mm per rotation, which could degrade the MVCT image quality. Each slice was reconstructed with half-gantry rotation, and the distance between reconstructed slices was 2 mm for fine mode, 4 mm for normal mode, and 6 mm for coarse mode. Acquired images were either manually or automatically registered with fusion planning CT data sets to determine the relative shifts for the patient repositioning before any actual treatment delivery. After performing the image verification, patient position was again verified with correct couch coordinates, and moved to the desired treatment location for treatment delivery. This system is equipped with xenon detectors similar to a CT scanner designed to obtain MVCT images for the image-guided radiation therapy usage of pretreatment setup verification and adaptive therapy. Our HT was installed in 2007; this is the first system in China and we have accumulated considerable clinical cases for comparison purpose. RapidArc system To compete with HT and other clinical advanced systems, the RA system has been studied and implemented by Varian Medical Systems to perform the VMAT treatment, which delivers treatment fractions possessing the characteristics of delivering single or multiple arcs.13 The technical advancement consisted of using varying dose rate, changing gantry rotation speed, and bidirectional MLC motion with variable MLC speed. This system consists of 40 leaf pairs at the center, each with a leaf width of 5 mm, and 20 leaf pairs on each side with a leaf width of 10 mm; therefore, the total number of leaves is 120 for a maximum field size of 40 40 cm2 treatment field. Volumetric image guidance is realized with kV cone-
beam CT guidance. The most significant differences between conventional IMRT and VMAT are the monitor units (MUs) required for delivering the prescribed dose per fraction and beam-on times from early study.14
Synergy S system The Synergy S system consists of a LINAC with step-and-shoot IMRT function and a high-resolution MLC with 40 leaf pairs with a leaf width of 4 mm (Beam modulator, Elekta, Crawley, UK).15 The S system is designed specifically for treating small field size targets with fine leaf resolution. Volumetric image guidance is performed with the cone-beam CT technique (Elekta XVI, Crawley, UK) and the standard segmental IMRT delivery. This platform has been implemented with the traditional segmental IMRT treatment at our facility since early 2000s.
Patients simulation and planning In all, 10 randomly selected patients with 12 isolated targets were under investigation. The involved vertebras in a target volume (TV) had only single-level, class I thoracic vertebral lesions, according to the spinal malignancy classification by DeWald et al.16 These patients were treated with CK using the Xsight Spine tracking system in our center as the baseline modality. Informed consents were obtained from all patients before receiving treatments. Among the selected lesions, 6 of the 12 lesions involved only vertebral body (body type) and 6 others involved vertebral body and unilateral pedicle (body þ pedicle type). Patients’ physical characteristics are summarized in Table 1. Key factors were patients’ KARNOFSKY Performance Status Z 70, life expectancy Z6 months, without symptomatic cord compression, vertebral compression fracture, epidural involvement, or prior spine irradiation. The minimum distance between the gross tumor and spinal cord (on magnetic resonance [MR] images) was Z2 mm. Before CK planning started, contrast-enhanced planning CT scanning (Brilliance CT Big Bore, Philips, Andover, MA) with 1.5-mm slice thickness and contrastenhanced 3 dimensional planning MR imaging (Discovery 1.5 T, GE, Milwaukee, WI) with FSPGR sequence of 1.2-mm slice thickness were performed in the supine treatment position for each patient. Planning CT and MR images were fused for gross tumor volume (GTV) and cord contouring. MR imaging with plain T2 sequence was also used for target definition. A treatment plan was generated using the CK MultiPlan 4.0.2 nonisocentric inverse-planning algorithm with tissue density heterogeneity corrections. As the CK beam was not flattened, 2 sets of collimators were used to maintain homogeneity. Typical beam and time reduction techniques were also implemented toward the end of the optimization process to produce a practically deliverable plan. Patients in the supine position were then treated by fitting into a vacuum cushion, taking advantages of fiducial-free spine tracking methodology. SBRT cases were also replanned with a HT 4.0.4 planning workstation for the HT system, with an Eclipse v10 workstation (Varian Medical Systems) for RA, and with Pinnacle3 8.0 (Philips Medical System) for 9 fields segmental IMRT. For the finest calculations, HT plans employed 1.0-cm jaw and 0.43-pitch combination to simulate the comparable plans from CK. Also RA plans employed dual arcs, from 179.91 to 180.11 counterclockwise and from 180.11 to 179.91 in the reverse direction to form 2 complete arc sweeps. The GTV was defined as the metastatic lesion within a vertebral body (body type) or within a vertebral body and its unilateral pedicle (body þ pedicle type). A safety margin expansion of 3 mm from GTV was contoured as the PTV. PTV was limited by the cord if overlapped, which was defined as the cord contoured from planning MR images plus an expansion of 2 mm. Cord and esophagus were extended 6 mm above and below PTV in craniocaudal direction, for accrual dosevolume histograms analysis.
Dosimetric evaluations Total treatment doses prescribed were 33 Gy/3F and 40 Gy/5F to PTV for bodytype and pedicle-type lesions, with a biologic effective dose (BED10) (α/β ¼ 10 Gy) of 69.3 Gy and 72 Gy, respectively. For target coverage (VPTV), 100% of the prescription dose line has to cover Z90% of the PTV. The dose-volume constraints for cord and esophagus also followed the protocol documented by Timmerman.17 Patients were then treated on consecutive days, with 1 fraction (F) per day until completion. The dose-volume histograms provided necessary information on dose distribution for comparing treatment plans with each modality. Based on these calculated histograms, gradient index (GI), dose heterogeneity index (HI), conformity index (CI), and beam-on time have been compiled for evaluation and comparison. The GI has been introduced to reflect the dose falloff outside the target. The low dose outside the prescription isodose volume (PIV) may cover significant amounts of normal tissues and can be responsible for complications, especially when the target is near critical structures. This index is defined as the ratio of the volume of half the prescription dose to the PIV.18 A low GI indicates a low dose spread outside the lesion and a rapid dose falloff. The Paddick’s CI factor,19 which takes into account the location of the PIV with respect to the TV, is defined as follows: CI ¼ (TV covered by PIV)2/(TV PIV). This
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
3
Table 1 Patients' clinical and dosimetric characteristics Patient Sex Age (y)
Primary tumor
Metastatic site
1 2 3
F M M
43 85 75
4 5 6-1
M M M
56 48 84
6-2
M
84
7-1
M
73
NSLC T10 NSLC T5 Renal adeno T7 carcinoma NSLC T9 NSLC T10 Prostate T7 adenocarcinoma Prostate T9 adenocarcinoma NSLC T3
7-2
M
73
NSLC
T5
8
M
65
T9
9
M
54
10
F
75
Rectal adenocarcinoma Stomach adenocarcinoma NSLC
T5 T5
Metastatic type
PTV volume (mm3)
VPTV (%) (CK)
VPTV (%) (HT)
VPTV (%) (RA)
VPTV (%) (IMRT)
Shortest distance PTV-cord (mm)
Shortest distance PTV-esophagus (mm)
Body Body Body
7427.22 23,483.56 23,956.25
95.2 94.2 90.5
95.1 91.7 92.1
96.2 91.9 93.5
91.7 87.8 88.7
1.18 0.62 0
9.1 3.3 2.0
Body Body Body
27,317.05 17,425.95 27,055.38
90.5 93.9 93.9
93.6 95.6 99.3
94.4 95.5 96.9
87.6 93.0 96.1
0 0 2.98
7.7 20 3.7
Body þ pedicle Body þ pedicle Body þ pedicle Body þ pedicle Body þ pedicle Body þ pedicle
19,990.05
95.5
94.9
98.6
90.7
3.24
12.6
4582.00
91.1
90.3
88.8
87.5
0
2.4
13,358.93
81.2
89.0
91.9
82.3
0
4.0
28,616.49
89.8
91.2
87.2
88.8
0
23.2
22,743.23
74.6
84.2
85.1
78.3
0
0
30,918.72
79.8
90.0
87.3
87.2
0
7.8
NSLC = non-small cell lung carcinoma. VPTV: PTV coverage.
n
index analyzes both the quality of the coverage and the volume of healthy tissue receiving a dose larger than or equal to that in the PIV. With the CI definition, a perfect plan would yield a CI equal to 1.00 and would be equal to the prescription isodose sculpted around the TV. The HI is defined as the ratio of the difference between the maximum and minimum dose to the mean dose of the PTV. This means that when the index is 0, the dose distribution to the PTV would be the most homogeneous, and increasing HI indicates increasing dose heterogeneity. Definition of V20% to 80% reflects the penumbra volume. The maximum dose (Dmax), minimum dose (Dmin), mean dose (Dmean), D50 (the dose received by 50% of the PTV volume), D98 (the dose received by 98% of the PTV volume) of the PTV, esophageal D5 (the dose received by 5% of the esophageal volume), cord D1 (the dose received by 1% of the cord volume), and total beam-on time of each dedicated treatment system were all reported and compared. The delivery dose rates of CK, HT, RA, and IMRT were 800, 900, 600, and 400 MU/min, respectively. Statistical analyses Various statistical analysis tools, such as 1-way analysis of variance test, KruskalWallis test, Bonferroni test, and LSD-t test (Statistical Package for the Social Sciences; IBM Corporation, NY) were implemented to evaluate the statistical significance of the differences between the different indices with all the systems. Therefore, p o 0.05 was considered to indicate statistical significance in our study cases.
Results Patients’ clinical and target characteristics and PTV coverage are presented in Table 1. Tables 2 and 3 summarize the dosimetric indices for body and body þ pedicle types of thoracic spinal lesions and each treatment modality. For the body-type lesions with a prescription dose of 33 Gy/3F to PTV, CK, HT, and RA satisfied VPTV Z 90% in all the 6 cases; however, IMRT reached this criterion in only 3 out of the 6 cases (50%). CK, HT, and RA achieved a sharp GI and a small penumbra volume compared to IMRT (p ¼ 0.001, 0.027, 0.01; p ¼ 0.002, 0.015, 0.001; respectively), without intergroup differences among CK, HT, and RA. RA achieved a comparable CI related to CK, HT, and IMRT (p ¼ 0.001, 0.047, 0.035, respectively), without intergroup differences among CK, HT, and IMRT. CK had a high Dmax, a high Dmean, and a high D50 in the PTV compared with HT, RA, and IMRT (p ¼ 0.000, 0.014, 0.015; p ¼ 0.001, 0.007, 0.002; p ¼ 0.010, 0.000, 0.000; respectively), reflecting a heterogeneous dose distribution in CK planning. The intergroup differences were significant (all with a p o 0.001) between any 2 of the 4 modalities for beam-on time and MU except between RA and IMRT, showing that CK had
the longest beam-on time and the largest MU, followed by HT, RA, and IMRT, which had the shortest beam-on time and the least amount of delivery time. For the body þ pedicle-type metastasis with a prescription dose of 40 Gy/5F to PTV, HT satisfied VPTV Z 90% in 4 of the 6 cases (66.7%), and reached a target coverage of 89.0% in another case. CK and RA reached VPTV Z 90% in 2 cases (33.3%), and only in 1 case (16.7%) by IMRT. As the results obtained for the body-type metastasis, CK, HT, and RA achieved a sharp GI and a small penumbra volume compared to IMRT (p ¼ 0.000, 0.000, 0.000; p ¼ 0.021, 0.066, 0.007; respectively), without intergroup differences among CK, HT, and RA. RA achieved a good CI compared with CK, HT, and IMRT (all had a p ¼ 0.000), without intergroup differences among CK, HT, and IMRT. The intergroup differences were significant (p ¼ 0.000 between CK and HT, p ¼ 0.035 between CK and RA, p ¼ 0.017 between CK and IMRT; p ¼ 0.000 between HT and RA, also between HT and IMRT) between any 2 of the 4 modalities for D50 in the PTV except between RA and IMRT, showing that CK had the highest figure of merit, followed by HT, RA, and IMRT, which had the lowest figure of merit. The intergroup differences were significant (all had p o 0.001) between any 2 of the 4 modalities for beam-on time except between HT and IMRT, and between RA and IMRT. The intergroup differences were significant (all had p o 0.001, except p ¼ 0.031 between HT and RA) between any 2 of the 4 modalities for MU except between HT and IMRT, and between RA and IMRT, showing that CK had the longest delivery time, followed by HT, RA, and IMRT, which had the shortest delivery time. A very typical dosimetry outcome can be observed in Fig. 1, and the difference between body-type and body þ pedicle-type TV is shown in Fig. 2.
Discussion From the past, 2-dimensional conventional radiation therapy has been a well-established palliative treatment paradigm for painful vertebral lesions, with a median duration of pain relief of approximately 3 to 6 months after irradiation. Although no difference was detected in pain response rate (from 50% to 90%) and dose-response relationship for pain control between single- and multifraction schedules in prospective studies, single fractionation
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
4
Table 2 Dosimetric comparison (33 Gy/3F to PTV) in body-type thoracic spinal metastases Dosimetric factor
CK
HT
RA
IMRT
PTV GI Mean SD
3.33 0.24
3.80 0.50
3.67 0.30
4.70 0.75
PTV CI Mean SD
0.63 0.08
0.79 0.14
0.98 0.22
0.77 0.17
PTV HI Mean SD
0.73 0.08
0.57 0.22
0.57 0.21
0.64 0.12
427.25 155.42
522.77 169.64
395.04 139.84
789.54 216.98
PTV Dmax (Gy) Mean SD
46.16 2.41
39.21 2.87
42.03 3.54
42.06 1.35
PTV Dmin (Gy) Mean SD
17.27 3.93
18.87 6.77
20.60 6.33
18.73 4.20
PTV Dmean (Gy) Mean SD
39.92 2.44
35.64 1.37
37.00 1.68
36.46 0.91
PTV D98 (Gy) Mean SD
25.10 5.66
28.21 3.21
30.20 2.52
26.58 4.03
PTV D50 (Gy) Mean SD
41.44 1.97
36.01 1.60
37.35 1.79
37.03 0.84
Beam-on time (min) Mean SD
55.33 6.95
29.17* 1.47
10.50 3.94
12.83 3.71
43,722.24 4048.97
24,993.50 1359.79
5785.00 1037.72
5077.43 1512.31
22.03 1.38
20.63 2.01
21.27 0.76
22.25 2.50
17.92 3.68
15.97 3.73
17.85 1.39
17.55 2.48
Esophagus Dmax (Gy) Mean SD
24.87 1.69
23.48 4.07
22.60 4.00
24.59 3.49
Esophagus D5 (Gy) Mean SD
18.75 1.32
17.69 3.25
17.43 3.44
15.56 6.02
PTV V20% Mean SD
to 80%
(mm3)
MU Mean SD Cord Dmax (Gy) Mean SD Cord D1 (Gy) Mean SD
F
p 12.322
0.006
4.925
0.010
1.133
0.360
6.432
0.003
6.916
0.002
0.375
0.772
7.239
0.002
1.771
0.185
13.295
0.000
128.523
0.000
374.833
0.000
1.026
0.402
0.568
0.643
0.460
0.714
0.429
0.934
SD ¼ standard deviation. n
In 2 times.
has been associated with a 20% to 25% incidence of repeat treatment vs 7% to 8% with fractionated radiation therapy. Patients treated with lower biologic effective dose were more frequently given second courses of radiation to the metastatic sites.20-23 Longcourse and higher dose radiation therapy for metastatic spinal
cord compression was significant for improved progression-free survival and local control.24 With the emerging targeted systemic therapies and much better understanding of cancer biology, patients might live longer with bony lesions. There is a good rationale for high dose escalation beyond palliative dose levels to
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
5
Table 3 Dosimetric comparisons (40 Gy/5F to PTV) for body þ pedicle-type thoracic spinal metastases Dosimetric factor
CK
HT
RA
IMRT
PTV GI Mean SD
3.23 0.22
3.58 0.43
3.95 0.35
6.33 1.32
PTV CI Mean SD
0.54 0.09
0.58 0.15
0.92 0.10
0.60 0.14
0.86 0.19
0.73 0.28
0.74 0.27
0.94 0.18
407.68 214.20
474.47 176.22
347.80 155.22
705.96 262.95
57.81 2.93
54.25 9.87
57.33 6.22
57.37 5.81
15.76 7.65
20.04 6.86
23.00 8.41
15.00 5.45
PTV Dmean (Gy) Mean SD
49.16 2.68
45.88 3.99
45.86 1.52
45.22 2.99
PTV D98 (Gy) Mean SD
26.60 6.68
29.65 5.44
31.97 5.29
25.80 5.91
PTV D50 (Gy) Mean SD
52.28 2.09
47.15 4.59
46.15 1.40
46.35 2.55
Beam-on time (min) Mean SD
61.67 8.14
20.17* 2.86
6.30 0.52
13.67 3.39
61,387.27 14527.16
17,258.00 2653.29
3807.00 326.10
5524.62 1373.21
Cord Dmax (Gy) Mean SD
31.00 4.27
28.49 1.55
28.78 1.36
27.12 1.88
Cord D1 (Gy) Mean SD
22.25 4.06
20.64 2.95
23.71 2.46
18.48 2.89
Esophagus Dmax (Gy) Mean SD
28.60 6.01
24.14 8.24
27.71 7.11
29.75 6.55
Esophagus D5 (Gy) Mean SD
22.95 6.24
19.47 7.59
21.73 5.84
23.62 8.16
PTV HI Mean SD V20% to 80% (mm3) Mean SD PTV Dmax (Gy) Mean SD PTV Dmin (Gy) Mean SD
MU Mean SD
n
F
p
17.257
0.001
11.941
0.000
1.095
0.374
3.466
0.036
0.363
0.781
1.643
0.211
2.219
0.117
1.416
0.268
37.664
0.000
21.161
0.000
21.147
0.000
5.340
0.149
3.056
0.052
0.596
0.627
0.338
0.798
In 2 times.
improve on existing rates of local and pain control, especially for those patients with solitary or oligo lesions and with a relatively indolent or well-controlled primary tumor, good performance status, as well as absence of visceral lesions. Recent advances in radiotherapy technology—stereotactic radiosurgery in generating sharp dose gradient, in combination with IMRT for highly conformal dose distributions and image tracking and guidance for
precise treatment delivery, have made curative-intent SBRT of spinal lesions possible and early results of pain and local tumor control were promising.25,26 The prospective study of Milano et al.8 concluded that bony oligo lesions and solitary metastasis from breast cancer were well controlled by the SBRT technique, with a fourfold and threefold reduced hazard of death, respectively. With our clinical planning evaluation of SBRT, to our best knowledge,
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
6
A
B
C
D
Fig. 1. Axial view of SBRT (40 Gy/5F) plan of a body þ pedicle-type metastasis (T9). PTV is contoured in red, cord in sky blue, and esophagus in white. (A) CyberKnife (VPTV ¼ 89.8%). (B) Helical tomotherapy (VPTV ¼91.2%). (C) RapidArc (VPTV ¼87.2%). (D) IMRT (VPTV ¼88.8%). Isodose lines: 50 Gy in black, 40 Gy in green, 30 Gy in yellow, 20 Gy in blue, and 10 Gy in pink. (Color version of figure is available online.)
this might be the first comparative dosimetric study on 4 different LINAC-based treatment modalities for isolated thoracic spinal metastasis SBRT.
For a curative-intent SBRT, high dose and conformal coverage of TV are critical. Yamada et al.27 suggested that radiation dose was a significant predictor of local control after SBRT for metastatic
A
B
C
D
E
F
Fig. 2. Illustration in 3 dimensions of a body-type metastasis (A-C) and a body þ pedicle-type metastasis (D-F). A margin of 3 mm from GTV (in green) is defined as PTV (in red). (Color version of figure is available online.)
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
spinal lesions. Koyfman et al.28 treated 208 spinal metastatic lesions on 149 patients with 10 to 16 Gy in a single fraction, and found that the incidence of marginal recurrence increased with a dose of o16 Gy. Choi et al.29 had treated 20 patients with gastrointestinal tract cancer with HT for 31 spinal oligo lesions, which showed that BED10 higher than 57 Gy to GTV could improve local control. However, an adverse event after spinal SBRT was vertebral compression fracture, whose risk increased significantly with a dose per fraction of 20 Gy or greater.30,31 Therefore, to summarize all the available clinical data, we chose to prescribe 33 Gy in 3F and 40 Gy in 5F to cover at least 90% of PTV volume, for body and body þ pedicle type, respectively. These prescriptions could translate the BED10 to 69.3 Gy and 72 Gy, correspondingly. In the process of treatment planning, as organs at risk (OARs) (e.g., cord and esophagus in this study) were protected with first priority, all the dosimetric indices for OARs showed no significant differences among the 4 modalities for sparing purposes in planning results. For the body-type lesions, CK, HT, and RA reached PTV coverage criterion in all the 6 cases; however, IMRT reached this criterion only in 50% of the cases. CK, HT, and RA also achieved a sharp GI and a small penumbra volume compared to IMRT, without intergroup differences among these 3 modalities. The clinical choice would be focused on the first 3 modalities. As RA achieved a good CI and at the same time had the shortest beam-on time and least MU compared with CK and HT, it would be the first choice for the body-type isolated spinal lesions. The second choice would be CK or HT. Although HT had a homogeneous dose distribution and shorter beam-on time compared with CK, a fraction of high-dose HT SBRT would be divided into 2 separated subfractions by the fractional dose constraint of 6 Gy in the HT system, which has no intrafraction target-tracking system like CK, so intrafractional movements and errors could take place during interfaction. To resolve this clinically relevant issue, one could consider a secondary MVCT image guidance before the second subfraction, which takes 3 to 5 minutes more. For the body þ pedicle-type lesions that had an irregular and large PTV volume compared with the body-type metastasis, HT completely satisfied target coverage in 4 of the 6 cases (66.7%) and reached 89.0% coverage of the PTV in another patient; CK and RA reached 90% coverage criterion in 2 cases (33.3%), and only in 1 case by IMRT. Like the case of body-type metastasis, CK, HT, and RA all achieved a sharp GI and a small penumbra volume compared to IMRT, without intergroup differences among the first 3 modalities. HT would be the preferential choice for the body þ pedicle-type isolated spinal metastasis, but at the cost of treatment time and MU compared with the RA technique. CK is a cone-based LINAC system that covered the lesion with nonisocentric noncoplanar circular beams spread over a large range of angles. With the use of multiple entries of different sizes, modulated beams, and the employment of inverse planning based on specific constraints, CK could achieve comparable conformity, while reducing significantly the treatment time compared with a gamma-knife system for benign intracranial lesions.32 Unlike the pretreatment image guidance of HT, RA, and Synergy S systems, CK is unique in that it uses intrafractional X-ray imaging (typically every 30 to 60 seconds), and automatic LINAC positional adjustments to compensate for any detected changes in target positioning. For relatively regular-shaped body-type metastasis, CK achieved high target coverage, sharp GI, and small penumbra volume like HT and RA, with a heterogeneous dose distribution, which however might not be a disadvantage as the hot spots were located within the TV. HT is a modality for delivering IMRT treatments using a rotating LINAC mounted on a continuously moving slip ring gantry in synchrony with the couch motion. This technique delivers highly conformal dose using 6-MV x-ray beams with a 64 leaves binary collimator of a 40-cm-
7
wide fan of thicknesses 1.0 to 5.0 cm to an isocenter 85 cm away from the source. The MLC leaves open 51 times per rotation as the gantry moves at a constant speed.10-12 In this dosimetric study of isolated thoracic spinal metastasis SBRT, with the high modulation resolution of its 1.0-cm jaw thickness, HT achieved a high target coverage even in complex-shaped body þ pedical-type lesions. In this study, the VMAT technique was accomplished by the RA system, which used more beam directions than fixed-gantry IMRT did and delivered highly conformal volumetric dose distributions in dual 3601 arcs. In addition to the marked advantage of its shorter treatment time and fewer MUs compared with other modalities of LINAC-based therapy, the combination of the advantages of arc therapy and IMRT, as demonstrated by Chin et al.33 in treatment planning with collimator motion, dose-rate variation, and various arc numbers, the highly conformal dose distributions could be achieved also for various target shapes, making RA a feasible choice for the body-type metastasis. In our results, RA could achieve relatively high target coverage, sharp GI, and small penumbra volume comparable to those of CK and HT, and also with an acceptable CI. However, the extra imaging time should be considered in comparing with similar technologies. The fine-leaf-width MLC in combination with IMRT technique could also yield dosimetric benefits in SBRT. Treatment of small lesions in cases involving complex target and OARs geometry would benefit from use of a fine-leaf-width MLC. In a previous dosimetric study for IMRT of spinal tumors, Wu et al.34 showed that the 2.5-mm leaf-width collimator significantly improved spinal cord sparing, with dose reductions of 14% to 19% in highto mid-level dose regions, compared with the 5-mm leaf width collimator. Chae et al.35 showed that smaller MLC leaf width provided improved target coverage in both IMRT and VMAT for spinal lesions, and its improvement was larger in IMRT than in VMAT. In addition, the smaller MLC leaf width was more effective for conforming to the complex-shaped targets. In this study, stepand-shoot IMRT-based SBRT was performed with the Synergy S system, which has a 4-mm leaf-width MLC. For small lesions, the limitation and the step-and-shoot beam path may not be efficient for spinal SBRT with our analysis, especially for the complexshaped body þ pedicle-type lesions, whose dose gradient requirement is a critical factor. In this case the current Synergy S system might not be able to meet the strict dosimetric requirements.
Conclusions All modalities are capable of performing SBRT based on the comparable GI and conformal index. Based on this dosimetric study of different LINAC-based SBRT modalities with isolated thoracic spinal metastasis, treatment delivery technique based on our planning results could be properly recommended: the RA is the first choice for the body-type metastasis owing to its efficiency (shortest treatment time) and similar dosimetric outcomes compared with other modalities; the second choice is CK or HT. However, HT is the preferential choice for the body þ pedicletype metastasis from its planning results. This study suggests further clinical investigation with clinical outcomes and systematic evaluations. As the results indicated, the step-and-shoot IMRT was clinically limited by its beam flexibility and dose gradient might not fulfill the dosimetric constraints in our study. Depending on the available technology, this conclusion was applied only to our center with current delivery systems addressed in the paper. Authors’ Contributions Ju.Y. participated in acquisition of data, performed the statistical analyses, and helped write the manuscript. L.M. and Ju.Y. were
8
J. Yang et al. / Medical Dosimetry ] (2016) ]]]–]]]
responsible for study conception and design, and drafted the manuscript. X.S.W., W.X., and X.H.C. contributed to data analyses. S.P.X. and Z.J.J. participated in the design and coordination of the study. L.D. and B.N.C. participated in acquisition of data. Ja.Y. reviewed and modified the contents and references. All authors read and approved the final manuscript. References 1. Gabriel, K.; Schiff, D. Metastatic spinal cord compression by solid tumors. Semin. Neurol. 24:375–83; 2004. 2. Schiff, D. Spinal cord compression. Neurol. Clin. 21:67–86; 2003. 3. Ryu, S.I.; Chang, S.D.; Kim, D.H.; et al. Image-guided hypo-fractionated stereotactic radiosurgery to spinal lesions. Neurosurgery 49:838–46; 2001. 4. Faul, C.M.; Flickinger, J.C. The use of radiation in the management of spinal lesions. J. Neurooncol. 23:149–61; 1995. 5. Sundaresan, N.; Digiacinto, G.V.; Hughes, J.E.O.; et al. Treatment of neoplastic spinal cord compression: Results of a prospective study. Neurosurgery 29:645–50; 1991. 6. Blomgren, H.; Lax, I.; Naslund, I.; et al. Stereotactic high dose fraction radiation therapy of extracranial tumors using an accelerator. Clinical experience of the first thirty-one patients. Acta Oncol. 34:861–70; 1995. 7. Hamilton, A.J.; Lulu, B.A.; Fosmire, H.; et al. Preliminary clinical experience with linear accelerator-based spinal stereotactic radiosurgery. Neurosurgery 36:311–9; 1995. 8. Milano, M.T.; Katz, A.W.; Zhang, H.; et al. Oligometastases treated with stereotactic body radiotherapy: long-term follow-up of prospective study. Int. J. Radiat. Oncol. Biol. Phys. 83:878–86; 2012. 9. Antypas, C.; Pantelis, E. Performance evaluation of a CyberKnife G4 imageguided robotic stereotactic radiosurgery system. Phys. Med. Biol. 53:4697–718; 2008. 10. Kestin, L.L.; Sharpe, M.B.; Frazier, R.C.; et al. Intensity modulation to improve dose uniformity with tangential breast radiotherapy: Initial clinical experience. Int. J. Radiat. Oncol. Biol. Phys. 48(5):1559–68; 2000. 11. Mackie, T.R.; Holmes, T.; Swerdloff, S.; et al. TomoTherapy: A new concept for the delivery of dynamic conformal radiotherapy. Med. Phys. 20:1709–19; 1993. 12. Jeraj, R.; Mackie, T.R.; Balog, J.; et al. Radiation characteristics of helical TomoTherapy. Med. Phys. 31:396–404; 2004. 13. Otto, K. Volumetric modulated arc therapy. IMRT in a single gantry arc. Med. Phys. 35:310–7; 2008. 14. Zhang, P.; Happersett, L.; Hunt, M.; et al. Volumetric modulated arc therapy: Planning and evaluation for prostate cancer cases. Int. J. Radiat. Oncol. Biol. Phys. 76(5):1456–62; 2010. 15. Guckenberger, M.; Sweeney, R.A.; Flickinger, J.C.; et al. Clinical practice of image-guided spine radiosurgery—Results from an international research consortium. Radiat. Oncol. 6:172–82; 2011. 16. Dewald, R.L.; Bridwell, K.H.; Prodromas, C.; et al. Reconstructive spinal surgery as palliation for metastatic malignancies of the spine. Spine 10:21–6; 1985. 17. Timmerman, R.D. An overview of hypofractionation and introduction to this issue of seminars in radiation oncology. Semin. Radiat. Oncol. 18:215–22; 2008.
18. Paddick, I.; Lippitz, B. A simple dose gradient measurement tool to complement the conformity index. J. Neurosurg. 105:194–201; 2006. 19. Paddick, I. A simple scoring ratio to index the conformity of radiosurgical treatment plans. J. Neurosurg. 93:219–22; 2000. 20. Chow, E.; Zeng, L.; Salvo, N.; et al. Update on the systemic review of palliative radiotherapy trials for bone metastasis. Clin. Oncol. (R. Coll. Radiol.) 24:112–24; 2012. 21. Lutz, S.; Berk, L.; Chang, E.; et al. American Society for Radiation Oncology (ASTRO). Palliative radiotherapy for bone lesions: An ASTRO evidence-based guideline. Int. J. Radiat. Oncol. Biol. Phys. 79:965–76; 2011. 22. Wu, J.S.; Wong, R.; Johnston, M.; et al. Meta-analysis of dose-fractionation radiotherapy trials for the palliation of painful bone metastases. Int. J. Radiat. Oncol. Biol. Phys. 55:594–605; 2003. 23. Chow, E.; Harris, K.; Fan, G.; et al. Palliative radiotherapy trials for bone lesions: A systematic review. J. Clin. Oncol. 25:1423–36; 2007. 24. Rades, D.; Lange, M.; Veninga, T.; et al. Preliminary results of spinal cord compression recurrence evaluation (score-1) study comparing short-course versus long-course radiotherapy for local control of malignant epidural spinal cord compression. Int. J. Radiat. Oncol. Biol. Phys. 73:228–34; 2009. 25. Joaquim, A.F.; Ghizoni, E.; Tedeschi, H.; et al. Stereotactic radiosurgery for spinal lesions: A literature review. Einstein (Sao Paulo) 11:247–55; 2013. 26. Sahgal, A.; Larson, D.A.; Chang, E.L. Stereotactic body radiosurgery for spinal lesions: A critical review. Int. J. Radiat. Oncol. Biol. Phys. 71:652–65; 2008. 27. Yamada, Y.; Bilsky, M.H.; Lovelock, D.M.; et al. High-dose, single fraction imageguided intensity-modulated radiotherapy for for metastatic spinal lesions. Int. J. Radiat. Oncol. Biol. Phys. 71:484–90; 2008. 28. Koyfman, S.A.; Djemil, T.; Burdick, M.J.; et al. Marginal recurrence requiring salvage radiotherapy after stereotactic body radiotherapy for spinal lesions. Int. J. Radiat. Oncol. Biol. Phys. 83:297–302; 2012. 29. Choi, Y.L.; Kim, J.W.; Lee, I.J.; et al. Helical TomoTherapy for spine oligometastases from gastrointestinal malignancies. Radiat. Oncol. J. 29:219–27; 2011. 30. Sahgal, A.L.; Atenafu, E.G.; Chao, S.; et al. Vertebral compression fracture after spine stereotactic body radiotherapy: A multi-institutional analysis with a focus on radiation dose and the spinal instability neoplastic score. J. Clin. Oncol. 31:3426–31; 2013. 31. Cunha, M.V.; Al-Omair, A.; Atenafu, E.G.; et al. Vertebral compression fracture (VCF) after spine stereotactic body radiation therapy (SBRT): Analysis of predictive factors. Int. J. Radiat. Oncol. Biol. Phys. 84:e343–9; 2012. 32. Gevaert, T.; Levivier, M.; Lacornerie, T.; et al. Dosimetric comparison of different treatment modalities for stereotactic radiosurgery of arteriovenous malformations and acoustic neuromas. Radiother. Oncol. 106:192–7; 2013. 33. Chin, L.M.; Kijewski, P.K.; Sevesson, G.K.; et al. Dose optimization with computer-controlled gantry rotation, collimator motion and dose-rate variation. Int. J. Radiat. Oncol. Biol. Phys. 9:723–9; 1983. 34. Wu, Q.,J.; Wang, Z.; Kirkpatrick, J.P.; et al. Impact of collimator leaf width and treatment technique on stereotactic radiosurgery and radiotherapy plans for intra- and extracranial lesions. Radiat. Oncol. 4:3–12; 2009. 35. Chae, S.M.; Lee, G.W.; Son, S.H. The effect of multileaf collimator leaf width on the radiosurgery planning for spine lesion treatment in terms of the modulated techniques and target complexity. Radiat. Oncol. 9:72–81; 2014.