- Email: [email protected]
Dosimetric comparison of metastatic spinal photon treatment techniques
Medical Dosimetry 37 (2012) 369-373
Medical Dosimetry journal homepage: www.meddos.org
Dosimetric comparison of metastatic spinal photon treatment techniques Marvene M. Ewing, B.S., C.M.D., Samuel M. Carnes, B.S., R.T.(T), Mark A. Henderson, M.D., and Indra J. Das, Ph.D., F.A.C.R., F.A.S.T.R.O. Department of Radiation Oncology, Indiana University School of Medicine, Indianapolis, IN
A R T I C L E
I N F O
Article history: Received 21 June 2011 Accepted 01 February 2012 Keywords: Dosimetry IMRT Spinal metastasis irradiation techniques
A B S T R A C T Traditional palliative treatment of metastatic cancer to the vertebral bodies often results in doses to the spinal cord that are higher than the dose prescribed to the target, or gross tumor volume (GTV). This study compares traditional techniques of spine palliation with intensity-modulated radiation therapy (IMRT). The purpose of the study is 2-fold: first, the study demonstrates the benefits of using IMRT to lower the dose to the organs at risk (OAR), particularly for the spinal cord and other nonspecified normal tissues; second, the article provides information regarding the advantages and disadvantages of commonly used conventional techniques for treating the vertebral bodies based on patient anatomy. Because the use of IMRT or other advanced techniques may be prohibitive because of insurance issues, treatment plans were created that compared optimal coverage vs. optimal sparing for single-field, wedged-pair, and opposedbeam arrangements. Fifty-five patients were selected and divided by location of target (cervical, thoracic, and lumbar spine) and also by the measured separation between the anterior and posterior surface of the patient at the level of mid-GTV. Within each anatomic category the patients again were divided into the categories of small, medium, and large based on separation. The patient dataset that most closely represented the average separation within each category was selected, resulting in a total of 9 patients, and the appropriate treatment plan techniques were calculated for each of the 9 patients. The results of the study do show that the use of IMRT is far superior when compared with other techniques, both for coverage and for sparing of the surrounding tissue, regardless of patient size and the section of spine being treated. Based on a combination of both target coverage and sparing of normal tissues, the conventional plan of choice may vary by both the section of spine to be treated and by the size of the patient. 䉷 2012 American Association of Medical Dosimetrists.
Introduction The spine is a common site for metastatic disease in the cancer patient. In the clinic, one frequently sees some patients return multiple times for new sites of spinal metastasis. Management of spinal cord compression requires a delicate balance between relieving the symptoms without complication, which has been discussed in literature.1– 4 Harel and Angelov5 showed that nearly 50% of metastatic cancers are in the spine and management is very diverse and includes surgery, isotope, chemotherapy, and radiation. Recently, ASTRO provided evidence-based guidelines for palliative radiotherapy of bone metastases.6 For radiation treatment of metastatic lesions, various modalities (i.e., 3D-conformal radiotherapy, intensity-modulated radiation therapy [IMRT], stereotactic body radiation therapy [SBRT], tomotherapy, CyberKnife) and radiation types as photon, electron,
Reprint requests to: Marvene M. Ewing, B.S., C.M.D., Indiana University Health, Department of Radiation Oncology, IU Simon Cancer Center, 535 Barnhill Dr., Indianapolis, IN 46202. E-mail: [email protected]
and protons are available.5,7-17 However, sometimes treatment options are limited due to the availability of resources or due to previous treatment being either at or close to the current site of disease. Spinal cord position is variable depending on body location18 and hence treatment techniques need to be variable. There are various approaches for treatment from a simple single field to extremely complex IMRT or SBRT plans. Kubo19 showed that for thoracic regions, opposed anteroposterior/posteroanterior (AP/PA) treatments can increase the spinal cord dose significantly. Dosimetry at various depths and the impact of the prescription was presented by Barton et al.20 The spinal cord has always been the most significant dose-limiting structure in treatment of the spine because of its proximity coupled with its radiation sensitivity. To spare the cord, various approaches, such as prone vs. supine position, PA, AP/PA, wedge pair, and IMRT have been attempted.12,21–23 Spinal treatments have also been attempted with electron beams; however, because of inhomogeneity correction and electron scatter at bone, such treatments are not universally accepted.9,11,13,24 Moreover, even high-energy electron beams do not achieve a sufficient depth dose to be effective.
0958-3947/$ – see front matter Copyright 䉷 2012 American Association of Medical Dosimetrists doi:10.1016/j.meddos.2012.02.002
370
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
Table 1 Characteristics of selected patients (small, medium, and large): spinal region and separation Small
Site Cervical Thoracic Lumbar
Medium
Large
Separation range
Number of datasets
Average separation
Separation range
Number of datasets
Average separation
Separation range
Number of datasets
Average separation
ⱕ10 cm ⱕ15 cm ⱕ15 cm
5 5 7
7.5 cm 12 cm 12.6 cm
⬎10–ⱕ 14 cm ⬎15–ⱕ 25 cm ⬎15–ⱕ 20 cm
7 7 6
12 cm 22.6 cm 18.4 cm
⬎14 cm ⬎25 cm ⬎20 cm
5 4 9
15.2 cm 28.4 cm 22.8 cm
In recent years, the radiation oncology community has been striving for more conformal doses, thereby achieving better tumor coverage and control while sparing more of the normal tissue and thus reducing side effects. A 2009 study25 demonstrated that a single PA field did not achieve the recommendation put forth in ICRU report 50,26 but that AP/PA plans did provide the coverage. This study compared 3 different treatment plans: (1) A single PA field using ICRU reference points, (2) a single PA using the International Bone Metastasis Consensus Working Party reference points, and (3) AP/PA using ICRU points. With improvement in treatment techniques, patients are surviving longer and hence long-term neurologic complications could be visible with poor treatments. Better dose homogeneity and the avoidance of over- and underdosing should be given importance in patient treatments with longer life expectancies.25 Advanced technologies like IMRT and CyberKnife are now an option to achieve conformal doses and avoid normal tissues, thus creating a higher therapeutic ratio.7 A systematic technical approach comparing various conventional treatment techniques is presented in this study. Although the topic is fairly simplistic in nature, it demonstrates the often unsatisfactory results for 3D conventional treatments of metastatic spine disease and also compares those with the results of IMRT planning for the same target volumes. Methods and Materials Patients were selected whose computed tomography scan covered 1 or all 3 spinal regions: cervical, thoracic and lumbar. Using the measuring tool in the ECLIPSE (Varian
Medical Systems, Palo Alto, CA) treatment planning system (TPS), the separation from anterior to posterior was measured at a point running directly through the middle of the spine for the cervical, thoracic, and lumbar spine at C5, T6, and L3, respectively, towing to skin-to-cord depth variations.18 To consolidate and summarize patient data, Table 1 demonstrates the range of separation, average separation, and number of datasets within each group as defined as small, medium, and large patients. The data tended to cluster around various common sizes, with there being exceptional sizes much larger or smaller than the rest of the grouping. One dataset for each site was selected as close to the midrange as possible within each category, thus representative of the size grouping. Nine datasets were selected. Thoracic and lumbar patients were all positioned prone and the cervical patients were either prone or supine. Target volumes were determined at the levels of C4-C6, T5-T7, and L2-L3, which included the entire body of the vertebrae while excluding the pedicles, spinous and transverse processes, and spinal cord. The field borders included one vertebral body superior and inferior to the target volume with the field edge set at the vertebral interspace and 1.5 cm laterally. The contoured volumes were designated as gross tumor volume (GTV)C, GTVT, and GTVL representing the cervical, thoracic, and lumber regions, respectively. Target volumes, described above, were drawn by the attending radiation oncologist who specializes in the central nervous system. Every target volume was drawn by the same radiation oncologist to avoid variations in the target design. Potential organs at risk (OARs) for each anatomic section were contoured. For the cervical region, these included thyroid, esophagus, parotids, spinal cord, brainstem, and unspecified normal tissue; in the thoracic region, the OAR included the esophagus, spinal cord, lungs, heart, stomach, liver, and unspecified normal tissue; and in the lumbar region, it included spinal cord, stomach, liver, kidneys, small bowel, and unspecified normal tissue. The unspecified tissue, or “normal tissue,” consisted of the entire body structure minus the contoured structures above and could be treated as remaining volume at risk (RVR) as defined by ICRU 83.27 In the thoracic and lumbar regions, all datasets included PA, AP/PA, and wedge-pair planning with combinations of 6- and 16-MV beams and using different beam weights, as well as 7-field IMRT. For all plans, the normalization point was placed at the isocenter,
Table 2 (a) Dosimetric summary of the GTV coverage and dose to various OARs for small, medium, and large patients with various treatment planned techniques for (a) cervical spine, (b) thoracic spine, and (c) lumber spine GTVC Max % Small separation: 8 cm PA 6x 102.8 PA 16x 102.6 AntWdgPr 6x 102.0 AntWdgPr 16x 101.5 LatSupObl 6x 101.6 LatSupObl 16x 100.8 7-Field IMRT 6x 106.4 Medium separation: 13 cm PA 6x 110.1 PA 16x 108.1 AntWdgPr 6x 104.9 AntWdgPr 16x 104.9 LatSupObl 6x 102.4 LatSupObl 16x 104.3 7 Field IMRT 6x 108.8 Large separation: 15 cm PA 6x 110.8 PA 16x 109.3 AntWdgPr 6x 107.6 AntWdgPr 16x 107.1 LatSupObl 6x 102.7 LatSupObl 16x 104.1 7 Field IMRT 6x 106.7
Spinal Cord V⬎105% (mL)
Parotids
Thyroid
Dmean %
Dmax %
Esophagus
Normal tissue
Dmean %
Dmax %
Dmax %
V⬎67 % (mL)
V95 %
Min %
Dmax %
98.7 100 100 100 100 100 100
94.1 96.4 95.9 96.9 97.4 98.5 98.6
109.6 104.3 98.8 101.1 99.9 100.2 96.8
4.5 0 0 0 0 0 0
1.0 ⬍1 1.4 1.7 8.3 7.4 ⬍1
93.6 96.3 102.5 95.0 103.7 99.7 65.0
88.3 91.9 96.6 80.2 86.8 80.9 33.8
94.2 97.0 102.4 98.1 103.1 100.5 93.8
115.9 104.5 103.1 101.2 103.6 100.1 96.8
180 169 133 120 157 147 11
86.1 97.3 99.5 100 100 99.1 100.0
88.4 92.5 94.5 95.1 95.0 98.2 99.0
116.5 114.1 102.2 101.6 103.4 104.7 98.1
11.5 11.3 0 0 0 0 0
3.7 3.7 14.4 12.9 19.2 16.8 1.9
91.4 94.2 103.5 103.0 101.5 99.7 70.0
80.1 84.7 88.9 88.1 62.9 61.8 30.0
90.9 94.8 100.9 101.5 99.3 99.4 95.8
138.4 123.5 109.3 105.6 104.6 104.8 97.7
655 674 468 449 449 430 43
79.3 94.1 99.5 100 88.4 98.4 99.8
86.2 91.9 94.0 96.5 90.0 93.8 95.0
120.6 116.2 100.0 104.5 103.5 104.2 97.4
12.6 13.0 0 0 0 0 0
1.5 ⬍1 1.2 1.1 3.3 3.6 ⬍1
90.6 92.7 111.3 105.7 101.0 100.6 69.9
61.9 66.0 73.5 66.8 60.8 60.4 27.2
91.0 93.8 110.6 104.5 100.2 100.1 97.8
150.2 129.5 111.9 107.5 108.1 104.7 101.9
620 665 347 300 711 635 52
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
371
Fig. 1. Dose coverage for cervical spine region. Lines represent medium dose coverage (95% isodose), location of maximum dose, and GTV for various techniques. (A) PA, (B) posterior wedge pair, (C) anterior wedge pair, and (D) IMRT.
which was the center of the GTV as described in ICRU 50.26 For the cervical region, plans were compared between techniques for PA, wedge pair, opposed laterals, and 7-field IMRT. The AP/PA beams were replaced by lateral beams that included a 15⬚ couch kick to avoid shoulders in the beam’s eye view. In the cervical spine region, anterior wedge-pair fields were also planned because they provided better coverage because of depth. In the lumbar and thoracic region, the wedge pair plans consisted of 2 posterior obliques with the angles optimized for each patient’s anatomy. Dynamic wedges were used for plan optimization. The IMRT plans used 7 equally spaced beams or 7 beams with separations of 40 – 60⬚ depending on the location of OAR structures. All plans were generated with the data from the Varian Trilogy treatment machine. The target coverage was evaluated by using maximum doses in the target, the volume covered by the prescription isodose line, the 95% isodose line based on 3D planning criteria from ICRU 50,26 and by the 90% isodose line. The 90% volume of GTV was included simply because in many cases the 95% isodose line did not cover the target volume.
the posterior wedge pair. In general, for the cervical spine, the anterior wedge pair resulted in optimal target coverage and minimized doses to critical structures and normal tissue. The lateral superior oblique had very good coverage but resulted in the highest dose to the parotid tissues. The dose received by the mandible, oral cavity, and parotids can change dramatically based on the neck position of the patient. The IMRT plans were the only plans that resulted in delivering a maximum dose to the spinal cord that was less than the prescribed dose to the tumor. Figure 1A–1D shows the comparison of the single PA spine, posterior wedge pair, anterior wedge pair, and IMRT.
Thoracic spine
Results Cervical spine The cervical spine data were evaluated for overall plan maximum dose, coverage of the target by the 100%, 95%, and 90% isodose lines; maximum and mean doses to the spinal cord, brainstem, thyroid, and esophagus, parotid glands, and thyroid; and also for sparing of the unspecified tissue, as shown in Table 2a. All of the conventional techniques resulted in limited dose coverage by the 100% isodose line, although several techniques had excellent coverage by the 95% isodose line, based on ICRU 50 criteria.26 For the cervical section of the spine, the anterior wedge pair resulted in improved target coverage and sparing of normal tissue over
The data for the thoracic spine plans again compared the overall plan maximum dose, coverage of the target by the 100%, 95%, and 90% isodose lines; maximum and mean dose to the spinal cord; sparing of unspecified tissues; and sparing of the lungs, stomach, heart, esophagus, and liver (Table 2b). Similar to the cervical spine plans, the best coverage and sparing of tissues was achieved with the IMRT plans. Figure 2A–2D shows dose distributions for the single PA field, posterior wedge pair, PA/AP, and IMRT, respectively. Again, all plans resulted in limited coverage by the 100% isodose line, but all conventional plans except for the single posterior field achieved adequate coverage by the 95% line. The lung was evaluated for both the mean dose and the percentage volume of V5. Lung doses
Table 2(b) GTVT Max% Small separation: 12 cm PA 6x 107.2 PA 16x 103.6 PostWdgPr 6x 103.4 PostWdgPr 16x 102.6 PA(6x):AP(16x) 2:1 104.5 weighting IMRT 6x 106.2 Medium separation: 22 cm PA 6x 112.4 PA 16x 110.4 PostWdgPr 6x 107.1 PostWdgPr 16x 107.3 PA(6x):AP(16x) 2:1 107.4 weighting IMRT 6x 105.6 Large separation: 28 cm PA 6x 116.4 PA 16x 114.2 PostWdgPr 6x 105.6 PostWdgPr 16x 106.0 111.9 PA(6x):AP(16x) 2:1 weighting IMRT 6x 104.1
Spinal cord
Lung
Heart Dmax %
Esophagus
Liver
Normal tissue
Dmean %
Dmax %
Dmax %
Dmax V⬎67%
V95 %
Min %
Dmax %
V⬎105% (mL)
D mean %
V⬎67 % (mL)
97 100 100 100 100
93.1 95.7 98.3 97.5 97.1
112.9 103.0 103.8 102.4 106.9
7.6 0 0 0 ⬍1
11.3 11.5 18.3 19.0 11.6
92.2 92.9 99.1 96.3 97.3
42.1 44.3 34.9 35.9 46.8
93.9 96.6 100.3 98.6 99.0
22.0 24.9 29.7 30.4 21.4
114.5 104.0 105.0 102.4 108.0
313 161 107 86 171
100
97
96.7
0
13.1
52.8
16.1
91.0
2.8
98.7
13
85.7 92.7 100 100 98.2
87.9 90.5 95.0 95.6 93.3
120.9 116.0 109.1 108.8 111.3
17.6 17 15.5 14.4 14.8
7.6 7.4 13.9 14.2 7.7
84.8 81.1 98.7 88.8 98.9
27.8 29.7 17.1 17.9 34.5
87.3 91.5 100.7 99.9 96.3
2.8 2.2 2.4 2.6 2.4
134.2 118.6 113.3 109.5 120.3
602 636 622 502 823
100
95.0
98.4
0
8.8
42.1
8.7
86.0
⬍1
97.4
59
81.6 89.8 100 100 100
85.8 89.3 95 97.2 98.4
121.7 118.5 106.5 107.0 111.2
16.4 15.9 ⬍1 5.5 14
10.4 10.3 17.4 17.1 10.8
84.1 84.2 103.6 101.8 104.7
26.8 29.5 24.6 26.9 40.2
86.4 90.8 101.1 101.1 97.3
85.1 82.5 104.2 98.1 106.7
142.8 125.7 110.4 107.7 123.2
780 817 826 670 1241
99.5
94.5
93.9
0
11.8
53.8
19.8
76.9
53.8
94.3
54
372
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
Fig. 2. Dose coverage for thoracic spine region. Lines represent medium dose coverage (95% isodose), location of maximum dose, and GTV for various techniques. (A) PA, (B) posterior wedge pair, (C) PA/AP mixed energy, and (D) IMRT.
are well within acceptable criteria regardless of the planning technique or the size of the patients. The sparing of critical structures met commonly used dose limits within our clinic when using the small patient dataset for most of the planning techniques. For the medium and large patient datasets, the posterior wedged pair had the most sparing for all structures excluding the lung. Equally weighted AP plans were generally unacceptable, faring the worst for heart, liver, normal tissues, and plan maximum doses. The unequally weighted AP plans did show improvement over the equally weighted plans but still had rather high plan maximums and heart dose maximums. The thoracic patient data seem to reflect better coverage and dose sparing when using the posterior wedge pair technique over the other conventional planning techniques. Lumbar spine The lumbar spine datasets were evaluated for overall plan maximum dose; target coverage by the 100%, 95%, and 90% isodose lines; maximum dose to the spinal cord; and volume doses to the small bowel, stomach, combined kidney, and normal tissue (Table 2c). Figure 3A–3D shows the dose distributions for single PA field, posterior wedge pair, PA/AP, and IMRT plans. As previously discussed for both the cervical and thoracic spines, the IMRT plans reflected marked improvement for the sparing of normal
tissues and critical structures over any of the conventional techniques. The coverage by the 95% isodose line was adequate with all techniques except for the single posterior field and the 6-MV posterior wedged pair. The small bowel dose increased with increasing patient separation, regardless of technique, but the combined kidney and stomach doses decreased with increasing separation. However, all were within acceptable dose tolerances as defined by QUANTEC.28 The small bowel dose was minimized most by using the posterior wedged pair technique, whereas the spinal cord and combined kidney doses were the lowest when using the parallel opposed AP fields. Discussion and Conclusion This paper demonstrates the superiority of the IMRT plans over conventional techniques for both coverage of the target volumes and the sparing of the normal tissue and critical structures. The IMRT plans resulted in a reduced volume of normal tissue receiving 67% of the prescription dose by factors ranging from 6.5–14 when compared with the conventional plans. When IMRT is not chosen as the method of treatment, the plan comparisons show that the commonly used single posterior spine port seldom covers the target volume with adequate dose as defined by ICRU 5026 and always delivers a higher dose to the spinal cord than to the vertebral body target. Maximum plan doses also often exceed 110% of the prescription dose.
Table 2(c) GTVL
Small separation: 12 cm PA 6x PA 16x PostWdgPr 6x PostWdgPr 16x PA(6x):AP(16x) 2:1 weighting IMRT 6x Medium separation: 18 cm PA 6x PA 16x PostWdgPr 6x PostWdgPr 16x PA(6x):AP(16x) 2:1 weighting IMRT 6x Large separation: 23 cm PA 6x PA 16x PostWdgPr 6x PostWdgPr 16x PA(6x):AP(16x) 2:1 weighting IMRT 6x
Spinal cord
Sm bowel
Combined kidney
Normal tissue
Max %
V95 %
Min %
Dmax %
V⬎10 % (mL)
Dmean %
Dmean %
Dmax V⬎67%
% cc
112.3 108.6 105.6 104.6 106
85.5 94.2 100 100 100
89.0 92.0 96 98.2 95.3
119.7 109.7 106.7 104.4 108.3
16.5 14.0 2.2 0 7.6
29.5 30.9 28.6 30.0 31.8
23.3 107.4 38.4 36.7 24.0
122.0 109.8 108.5 105.5 109.9
429 448 346 285 442
104.7
100
95.2
98.4
0
13.0
10.6
98.7
60
116.1 114.7 109.6 106.2 107.3
76.2 89.0 87.3 100 100
86.0 90 90.4 97.2 95.1
130.6 127.3 114.2 106.9 113.5
35.5 34.8 28.3 5.9 31.5
33.4 34.8 33.3 34.9 36.7
6.3 5.1 13.0 11.6 6.4
156.5 136.5 125.4 108.1 126.9
1353 1542 1780 1519 1661
107.2
100
95
97.0
0
13.4
7.1
98.5
132
119.2 117.3 115.6 107.1 109
75.1 86.1 845.2 100 99.7
85.9 89.0 88.8 97.1 94.2
137.0 132.9 125.9 108.2 116.3
26.4 25.8 24.1 9.0 23.5
14.4 14.9 16.0 16.7 16.4
3.9 2.6 7.4 5.9 3.7
173.4 148.8 139.2 109.4 135.4
1527 1618 2170 1973 2063
104.5
100
96
97.7
0
7.7
14.7
98.8
158
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
373
Fig. 3. Dose coverage for lumber spine region. Lines represent medium dose coverage (95% isodose), location of maximum dose, and GTV for various techniques. (A) PA, (B) posterior wedge pair, (C) PA/AP mixed energy, and (D) IMRT.
For the cervical spine patients, the plan comparisons show that the anterior wedge pair meets more of the desired criteria than any of the other non-IMRT plans. The thoracic conventional plans demonstrate comparable coverage of the target by either the posterior wedge pair or parallel opposed AP with optimized weighting. Both of those planning techniques give similar doses to the spinal cord and those maximum doses ranged from 102–111% of the prescribed dose. The AP field combination gives the best sparing to the lung, but the posterior wedge pair is superior at sparing the stomach, heart, esophagus, liver, and normal tissues. The lumbar spine conventional planning techniques illustrate excellent coverage by the 95% isodose line for all but the single posterior field, and in some cases, by any plan using 6 MV rather than a higher energy. The spinal cord maximum doses are best tailored by using the AP optimized weighted beams over the posterior wedge pair, but there was very little difference among the mean doses to the spinal cord. The small bowel and stomach doses were best minimized by using the posterior wedge pair, whereas the combined kidney doses were the lowest by using the AP technique. The IMRT technique is shown to be superior to any of the other techniques but may not be the most practical for treating the metastatic patient who is in pain and cannot lie still for a long period of time. However, it does allow the doses to critical structures and normal tissues to be better controlled and it could allow the typical palliative doses to be increased, or allow “retreatment” of the same vertebral body. The technique to be used is ultimately chosen at the discretion of the treating physician, but these data do demonstrate the superiority of certain techniques based on the anatomic section of the spine to be treated and the separation of the patient. The resulting data from this exercise could be used by the treatment planner or a physician in determining which treatment technique might be better suited for their patient, based on the region of the spine involved and the size of the patient. This study could be further extended to advanced technologies and techniques, such as IMRT, SBRT, tomotherapy,10 and protons.15 We did not include the use of proton therapy (even though it is available at our center) because it is not readily available to most patients or clinics. However, it may be advantageous in many patients especially in pediatric patients.16,17,29 One should not become complacent merely because a treatment technique is considered adequate. The goal is to deliver the best treatment possible and to always strive for continuous quality improvement. References 1. Harries B. Spinal cord compression. BMJ. 1:611– 4; 1970. 2. Findlay G.F. Adverse effects of the management of malignant spinal cord compression. J. Neurol. Neurosurg. Psychiatry. 47:761– 8; 1984. 3. Harries B. Spinal cord compression. II. BMJ. 1:673– 6; 1970. 4. Windeyer B. Metastases in the central nervous system: treatment by radiotherapy and chemotherapy. Proc. Royal. Soc. Med. 57:1153–9; 1964.
5. Harel R. Angelov L. Spine metastases: current treatments and future directions. Eur. J. Cancer. 46:2696 –707; 2010. 6. Lutz S.; Berk L.; Chang E.; et al. Palliative radiotherapy for bone metastases: an ASTRO evidence-based guideline. Int. J. Radiat. Oncol. Biol. Phys. 79:965–76; 2011. 7. Chang U.K.; Youn S.M.; Park S.Q.; et al. Clinical results of cyberknife radiosurgery for spinal metastases. J. Korean. Neurosurg. Soc. 46:538 – 44; 2009. 8. Gong Y.; Wang J.; Bai S.; et al. Conventionally fractionated image-guided intensity modulated radiotherapy (IG-IMRT): a safe and effective treatment for cancer spinal metastasis. Radiol. Oncol. 11:1–10; 2008. 9. Li C.; Muller-Runkel R.; Vijayakumar S.; et al. Craniospinal axis irradiation: an improved electron technique for irradiation of the spinal axis. Br. J. Radiol. 67:186 –93; 1994. 10. McIntosh A.; Dunlap N.; Sheng K.; et al. Helical tomotherapy-based STAT RT: dosimetric evaluation for clinical implementation of a rapid radiation palliation program. Med. Dosimetry. 35:280 – 6; 2010. 11. Muller-Runkel R.; Vijayakumar, S. Spinal axis irradiation with electrons: measurements of attenuation by the spinal processes. Med. Phys. 13:539 – 43; 1986. 12. Nutting C.; Convery D.; Cosgrove V.; et al. Improvements in target coverage and reduced spinal cord irraditaion using intensity-modulated radiotherapy (IMRT) in patients with carcinoma of the thyroid gland. Radiother. Oncol. 60:173– 80; 2001. 13. Roback D.M.; Johnson J.M.; Khan F.M.; et al. The use of tertiary collimation for spinal irradiation with extended SSD electron fields. Int. J. Radiat. Oncol. Biol. Phys. 37: 1187–92; 1997. 14. Wowra B.; Muacevic A.; Zausinger S.; et al. Radiosurgery for spinal malignant tumors. Deutsches ãrzteblatt Internationaux 106:106 –12; 2009. 15. St. Clair W.H.; Adams J.A.; Bues M.; et al. Advantage of protons compared to conventional X-ray or IMRT in the treatment of a pediatric patient with medulloblastoma. Int. J. Radiat. Oncol. Biol. Phys. 58:727–34; 2004. 16. DeLaney T.F.; Liebsch N.J.; Pedlow F.X.; et al. Phase II study of high-dose photon/ proton radiotherapy in the management of spine sarcomas. Int. J. Radiat. Oncol. Biol. Phys. 74:732–9; 2009. 17. Yuh G.E.; Loredo L.N.; Yonemoto L.T.; et al. Reducing toxicity from craniospinal irradiation: using proton beams to treat medulloblastoma in young children. Cancer. J. 10:386 –90; 2004. 18. Brinkley D.; Masters, H.E. The depth of the spinal cord below the skin. Br. J. Radiol. 40:66 – 8; 1967. 19. Kubo H. Estimate of the excessive spinal cord dose in AP-PA thorax treatment. Radiother. Oncol. 13:187–92; 1989. 20. Barton R.; Robinson G.; Gutierrez E.; et al. Palliative radiation for vertebral metastases: the effect of variation in prescription parameters on the dose received at depth. Int. J. Radiat. Oncol. Biol. Phys. 52:1083–91; 2001. 21. Vijayakumar S. Treatment of metastatic spinal cord compression: an unsettled issue. J. Nat. Med. Assoc. 75:297–303; 1984. 22. Vijayakumar S.; Muller-Runkel, R. Irradiation of the thoracic esophagus. Prone versus supine treatment positions. Acta. Radiol. Oncol. 25:187–9; 1986. 23. Yin F.F.; Ryu S.; Ajlouni M.; et al. A technique of intensity-modulated radiosurgery (IMRS) for spinal tumors. Med. Phys. 29:2815–22; 2002. 24. Endicott T.; Fisher B.; Wong E.; et al. Pulmonary sequelae after electron spinal irradiation. Radiother. Oncol. 60:267–72; 2001. 25. Andic F.; Baz Cifci S.; Ors Y.; et al. A dosimetric comparison of different treatment plans of palliative spinal bone irradiation: analysis of dose coverage with respect to ICRU 50 report. J. Exp. Clin. Cancer. Res. 28:2; 2009. 26. ICRU Report 50. Prescribing, Recording, and reporting Photon Beam Therapy. Bethesda, MD: International Commission on Radiation Units and Measurements; 1993. 27. ICRU Report 83. In Prescribing, Recording, and Reporting Intensity-Modulated PhotonBeam Therapy. Bethesda, MD: International Commission on Radiation Units and Measurements; 2010. 28. Marks L.B.; Ten Haken R.K.; Martel M.K. Quantitative analyses of normal tissue effects in the clinic (QUANTEC). Int. J. Radiat. Oncol. Biol. Phys. 76;(suppl):S1–S160; 2010. 29. Athar B.S.; Paganetti, H. Comparison of second cancer risk due to out-of-field doses from 6-MV IMRT and proton therapy based on 6 pediatric patient treatment plans. Radiother. Oncol. 98:87–92; 2011.
Medical Dosimetry journal homepage: www.meddos.org
Dosimetric comparison of metastatic spinal photon treatment techniques Marvene M. Ewing, B.S., C.M.D., Samuel M. Carnes, B.S., R.T.(T), Mark A. Henderson, M.D., and Indra J. Das, Ph.D., F.A.C.R., F.A.S.T.R.O. Department of Radiation Oncology, Indiana University School of Medicine, Indianapolis, IN
A R T I C L E
I N F O
Article history: Received 21 June 2011 Accepted 01 February 2012 Keywords: Dosimetry IMRT Spinal metastasis irradiation techniques
A B S T R A C T Traditional palliative treatment of metastatic cancer to the vertebral bodies often results in doses to the spinal cord that are higher than the dose prescribed to the target, or gross tumor volume (GTV). This study compares traditional techniques of spine palliation with intensity-modulated radiation therapy (IMRT). The purpose of the study is 2-fold: first, the study demonstrates the benefits of using IMRT to lower the dose to the organs at risk (OAR), particularly for the spinal cord and other nonspecified normal tissues; second, the article provides information regarding the advantages and disadvantages of commonly used conventional techniques for treating the vertebral bodies based on patient anatomy. Because the use of IMRT or other advanced techniques may be prohibitive because of insurance issues, treatment plans were created that compared optimal coverage vs. optimal sparing for single-field, wedged-pair, and opposedbeam arrangements. Fifty-five patients were selected and divided by location of target (cervical, thoracic, and lumbar spine) and also by the measured separation between the anterior and posterior surface of the patient at the level of mid-GTV. Within each anatomic category the patients again were divided into the categories of small, medium, and large based on separation. The patient dataset that most closely represented the average separation within each category was selected, resulting in a total of 9 patients, and the appropriate treatment plan techniques were calculated for each of the 9 patients. The results of the study do show that the use of IMRT is far superior when compared with other techniques, both for coverage and for sparing of the surrounding tissue, regardless of patient size and the section of spine being treated. Based on a combination of both target coverage and sparing of normal tissues, the conventional plan of choice may vary by both the section of spine to be treated and by the size of the patient. 䉷 2012 American Association of Medical Dosimetrists.
Introduction The spine is a common site for metastatic disease in the cancer patient. In the clinic, one frequently sees some patients return multiple times for new sites of spinal metastasis. Management of spinal cord compression requires a delicate balance between relieving the symptoms without complication, which has been discussed in literature.1– 4 Harel and Angelov5 showed that nearly 50% of metastatic cancers are in the spine and management is very diverse and includes surgery, isotope, chemotherapy, and radiation. Recently, ASTRO provided evidence-based guidelines for palliative radiotherapy of bone metastases.6 For radiation treatment of metastatic lesions, various modalities (i.e., 3D-conformal radiotherapy, intensity-modulated radiation therapy [IMRT], stereotactic body radiation therapy [SBRT], tomotherapy, CyberKnife) and radiation types as photon, electron,
Reprint requests to: Marvene M. Ewing, B.S., C.M.D., Indiana University Health, Department of Radiation Oncology, IU Simon Cancer Center, 535 Barnhill Dr., Indianapolis, IN 46202. E-mail: [email protected]
and protons are available.5,7-17 However, sometimes treatment options are limited due to the availability of resources or due to previous treatment being either at or close to the current site of disease. Spinal cord position is variable depending on body location18 and hence treatment techniques need to be variable. There are various approaches for treatment from a simple single field to extremely complex IMRT or SBRT plans. Kubo19 showed that for thoracic regions, opposed anteroposterior/posteroanterior (AP/PA) treatments can increase the spinal cord dose significantly. Dosimetry at various depths and the impact of the prescription was presented by Barton et al.20 The spinal cord has always been the most significant dose-limiting structure in treatment of the spine because of its proximity coupled with its radiation sensitivity. To spare the cord, various approaches, such as prone vs. supine position, PA, AP/PA, wedge pair, and IMRT have been attempted.12,21–23 Spinal treatments have also been attempted with electron beams; however, because of inhomogeneity correction and electron scatter at bone, such treatments are not universally accepted.9,11,13,24 Moreover, even high-energy electron beams do not achieve a sufficient depth dose to be effective.
0958-3947/$ – see front matter Copyright 䉷 2012 American Association of Medical Dosimetrists doi:10.1016/j.meddos.2012.02.002
370
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
Table 1 Characteristics of selected patients (small, medium, and large): spinal region and separation Small
Site Cervical Thoracic Lumbar
Medium
Large
Separation range
Number of datasets
Average separation
Separation range
Number of datasets
Average separation
Separation range
Number of datasets
Average separation
ⱕ10 cm ⱕ15 cm ⱕ15 cm
5 5 7
7.5 cm 12 cm 12.6 cm
⬎10–ⱕ 14 cm ⬎15–ⱕ 25 cm ⬎15–ⱕ 20 cm
7 7 6
12 cm 22.6 cm 18.4 cm
⬎14 cm ⬎25 cm ⬎20 cm
5 4 9
15.2 cm 28.4 cm 22.8 cm
In recent years, the radiation oncology community has been striving for more conformal doses, thereby achieving better tumor coverage and control while sparing more of the normal tissue and thus reducing side effects. A 2009 study25 demonstrated that a single PA field did not achieve the recommendation put forth in ICRU report 50,26 but that AP/PA plans did provide the coverage. This study compared 3 different treatment plans: (1) A single PA field using ICRU reference points, (2) a single PA using the International Bone Metastasis Consensus Working Party reference points, and (3) AP/PA using ICRU points. With improvement in treatment techniques, patients are surviving longer and hence long-term neurologic complications could be visible with poor treatments. Better dose homogeneity and the avoidance of over- and underdosing should be given importance in patient treatments with longer life expectancies.25 Advanced technologies like IMRT and CyberKnife are now an option to achieve conformal doses and avoid normal tissues, thus creating a higher therapeutic ratio.7 A systematic technical approach comparing various conventional treatment techniques is presented in this study. Although the topic is fairly simplistic in nature, it demonstrates the often unsatisfactory results for 3D conventional treatments of metastatic spine disease and also compares those with the results of IMRT planning for the same target volumes. Methods and Materials Patients were selected whose computed tomography scan covered 1 or all 3 spinal regions: cervical, thoracic and lumbar. Using the measuring tool in the ECLIPSE (Varian
Medical Systems, Palo Alto, CA) treatment planning system (TPS), the separation from anterior to posterior was measured at a point running directly through the middle of the spine for the cervical, thoracic, and lumbar spine at C5, T6, and L3, respectively, towing to skin-to-cord depth variations.18 To consolidate and summarize patient data, Table 1 demonstrates the range of separation, average separation, and number of datasets within each group as defined as small, medium, and large patients. The data tended to cluster around various common sizes, with there being exceptional sizes much larger or smaller than the rest of the grouping. One dataset for each site was selected as close to the midrange as possible within each category, thus representative of the size grouping. Nine datasets were selected. Thoracic and lumbar patients were all positioned prone and the cervical patients were either prone or supine. Target volumes were determined at the levels of C4-C6, T5-T7, and L2-L3, which included the entire body of the vertebrae while excluding the pedicles, spinous and transverse processes, and spinal cord. The field borders included one vertebral body superior and inferior to the target volume with the field edge set at the vertebral interspace and 1.5 cm laterally. The contoured volumes were designated as gross tumor volume (GTV)C, GTVT, and GTVL representing the cervical, thoracic, and lumber regions, respectively. Target volumes, described above, were drawn by the attending radiation oncologist who specializes in the central nervous system. Every target volume was drawn by the same radiation oncologist to avoid variations in the target design. Potential organs at risk (OARs) for each anatomic section were contoured. For the cervical region, these included thyroid, esophagus, parotids, spinal cord, brainstem, and unspecified normal tissue; in the thoracic region, the OAR included the esophagus, spinal cord, lungs, heart, stomach, liver, and unspecified normal tissue; and in the lumbar region, it included spinal cord, stomach, liver, kidneys, small bowel, and unspecified normal tissue. The unspecified tissue, or “normal tissue,” consisted of the entire body structure minus the contoured structures above and could be treated as remaining volume at risk (RVR) as defined by ICRU 83.27 In the thoracic and lumbar regions, all datasets included PA, AP/PA, and wedge-pair planning with combinations of 6- and 16-MV beams and using different beam weights, as well as 7-field IMRT. For all plans, the normalization point was placed at the isocenter,
Table 2 (a) Dosimetric summary of the GTV coverage and dose to various OARs for small, medium, and large patients with various treatment planned techniques for (a) cervical spine, (b) thoracic spine, and (c) lumber spine GTVC Max % Small separation: 8 cm PA 6x 102.8 PA 16x 102.6 AntWdgPr 6x 102.0 AntWdgPr 16x 101.5 LatSupObl 6x 101.6 LatSupObl 16x 100.8 7-Field IMRT 6x 106.4 Medium separation: 13 cm PA 6x 110.1 PA 16x 108.1 AntWdgPr 6x 104.9 AntWdgPr 16x 104.9 LatSupObl 6x 102.4 LatSupObl 16x 104.3 7 Field IMRT 6x 108.8 Large separation: 15 cm PA 6x 110.8 PA 16x 109.3 AntWdgPr 6x 107.6 AntWdgPr 16x 107.1 LatSupObl 6x 102.7 LatSupObl 16x 104.1 7 Field IMRT 6x 106.7
Spinal Cord V⬎105% (mL)
Parotids
Thyroid
Dmean %
Dmax %
Esophagus
Normal tissue
Dmean %
Dmax %
Dmax %
V⬎67 % (mL)
V95 %
Min %
Dmax %
98.7 100 100 100 100 100 100
94.1 96.4 95.9 96.9 97.4 98.5 98.6
109.6 104.3 98.8 101.1 99.9 100.2 96.8
4.5 0 0 0 0 0 0
1.0 ⬍1 1.4 1.7 8.3 7.4 ⬍1
93.6 96.3 102.5 95.0 103.7 99.7 65.0
88.3 91.9 96.6 80.2 86.8 80.9 33.8
94.2 97.0 102.4 98.1 103.1 100.5 93.8
115.9 104.5 103.1 101.2 103.6 100.1 96.8
180 169 133 120 157 147 11
86.1 97.3 99.5 100 100 99.1 100.0
88.4 92.5 94.5 95.1 95.0 98.2 99.0
116.5 114.1 102.2 101.6 103.4 104.7 98.1
11.5 11.3 0 0 0 0 0
3.7 3.7 14.4 12.9 19.2 16.8 1.9
91.4 94.2 103.5 103.0 101.5 99.7 70.0
80.1 84.7 88.9 88.1 62.9 61.8 30.0
90.9 94.8 100.9 101.5 99.3 99.4 95.8
138.4 123.5 109.3 105.6 104.6 104.8 97.7
655 674 468 449 449 430 43
79.3 94.1 99.5 100 88.4 98.4 99.8
86.2 91.9 94.0 96.5 90.0 93.8 95.0
120.6 116.2 100.0 104.5 103.5 104.2 97.4
12.6 13.0 0 0 0 0 0
1.5 ⬍1 1.2 1.1 3.3 3.6 ⬍1
90.6 92.7 111.3 105.7 101.0 100.6 69.9
61.9 66.0 73.5 66.8 60.8 60.4 27.2
91.0 93.8 110.6 104.5 100.2 100.1 97.8
150.2 129.5 111.9 107.5 108.1 104.7 101.9
620 665 347 300 711 635 52
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
371
Fig. 1. Dose coverage for cervical spine region. Lines represent medium dose coverage (95% isodose), location of maximum dose, and GTV for various techniques. (A) PA, (B) posterior wedge pair, (C) anterior wedge pair, and (D) IMRT.
which was the center of the GTV as described in ICRU 50.26 For the cervical region, plans were compared between techniques for PA, wedge pair, opposed laterals, and 7-field IMRT. The AP/PA beams were replaced by lateral beams that included a 15⬚ couch kick to avoid shoulders in the beam’s eye view. In the cervical spine region, anterior wedge-pair fields were also planned because they provided better coverage because of depth. In the lumbar and thoracic region, the wedge pair plans consisted of 2 posterior obliques with the angles optimized for each patient’s anatomy. Dynamic wedges were used for plan optimization. The IMRT plans used 7 equally spaced beams or 7 beams with separations of 40 – 60⬚ depending on the location of OAR structures. All plans were generated with the data from the Varian Trilogy treatment machine. The target coverage was evaluated by using maximum doses in the target, the volume covered by the prescription isodose line, the 95% isodose line based on 3D planning criteria from ICRU 50,26 and by the 90% isodose line. The 90% volume of GTV was included simply because in many cases the 95% isodose line did not cover the target volume.
the posterior wedge pair. In general, for the cervical spine, the anterior wedge pair resulted in optimal target coverage and minimized doses to critical structures and normal tissue. The lateral superior oblique had very good coverage but resulted in the highest dose to the parotid tissues. The dose received by the mandible, oral cavity, and parotids can change dramatically based on the neck position of the patient. The IMRT plans were the only plans that resulted in delivering a maximum dose to the spinal cord that was less than the prescribed dose to the tumor. Figure 1A–1D shows the comparison of the single PA spine, posterior wedge pair, anterior wedge pair, and IMRT.
Thoracic spine
Results Cervical spine The cervical spine data were evaluated for overall plan maximum dose, coverage of the target by the 100%, 95%, and 90% isodose lines; maximum and mean doses to the spinal cord, brainstem, thyroid, and esophagus, parotid glands, and thyroid; and also for sparing of the unspecified tissue, as shown in Table 2a. All of the conventional techniques resulted in limited dose coverage by the 100% isodose line, although several techniques had excellent coverage by the 95% isodose line, based on ICRU 50 criteria.26 For the cervical section of the spine, the anterior wedge pair resulted in improved target coverage and sparing of normal tissue over
The data for the thoracic spine plans again compared the overall plan maximum dose, coverage of the target by the 100%, 95%, and 90% isodose lines; maximum and mean dose to the spinal cord; sparing of unspecified tissues; and sparing of the lungs, stomach, heart, esophagus, and liver (Table 2b). Similar to the cervical spine plans, the best coverage and sparing of tissues was achieved with the IMRT plans. Figure 2A–2D shows dose distributions for the single PA field, posterior wedge pair, PA/AP, and IMRT, respectively. Again, all plans resulted in limited coverage by the 100% isodose line, but all conventional plans except for the single posterior field achieved adequate coverage by the 95% line. The lung was evaluated for both the mean dose and the percentage volume of V5. Lung doses
Table 2(b) GTVT Max% Small separation: 12 cm PA 6x 107.2 PA 16x 103.6 PostWdgPr 6x 103.4 PostWdgPr 16x 102.6 PA(6x):AP(16x) 2:1 104.5 weighting IMRT 6x 106.2 Medium separation: 22 cm PA 6x 112.4 PA 16x 110.4 PostWdgPr 6x 107.1 PostWdgPr 16x 107.3 PA(6x):AP(16x) 2:1 107.4 weighting IMRT 6x 105.6 Large separation: 28 cm PA 6x 116.4 PA 16x 114.2 PostWdgPr 6x 105.6 PostWdgPr 16x 106.0 111.9 PA(6x):AP(16x) 2:1 weighting IMRT 6x 104.1
Spinal cord
Lung
Heart Dmax %
Esophagus
Liver
Normal tissue
Dmean %
Dmax %
Dmax %
Dmax V⬎67%
V95 %
Min %
Dmax %
V⬎105% (mL)
D mean %
V⬎67 % (mL)
97 100 100 100 100
93.1 95.7 98.3 97.5 97.1
112.9 103.0 103.8 102.4 106.9
7.6 0 0 0 ⬍1
11.3 11.5 18.3 19.0 11.6
92.2 92.9 99.1 96.3 97.3
42.1 44.3 34.9 35.9 46.8
93.9 96.6 100.3 98.6 99.0
22.0 24.9 29.7 30.4 21.4
114.5 104.0 105.0 102.4 108.0
313 161 107 86 171
100
97
96.7
0
13.1
52.8
16.1
91.0
2.8
98.7
13
85.7 92.7 100 100 98.2
87.9 90.5 95.0 95.6 93.3
120.9 116.0 109.1 108.8 111.3
17.6 17 15.5 14.4 14.8
7.6 7.4 13.9 14.2 7.7
84.8 81.1 98.7 88.8 98.9
27.8 29.7 17.1 17.9 34.5
87.3 91.5 100.7 99.9 96.3
2.8 2.2 2.4 2.6 2.4
134.2 118.6 113.3 109.5 120.3
602 636 622 502 823
100
95.0
98.4
0
8.8
42.1
8.7
86.0
⬍1
97.4
59
81.6 89.8 100 100 100
85.8 89.3 95 97.2 98.4
121.7 118.5 106.5 107.0 111.2
16.4 15.9 ⬍1 5.5 14
10.4 10.3 17.4 17.1 10.8
84.1 84.2 103.6 101.8 104.7
26.8 29.5 24.6 26.9 40.2
86.4 90.8 101.1 101.1 97.3
85.1 82.5 104.2 98.1 106.7
142.8 125.7 110.4 107.7 123.2
780 817 826 670 1241
99.5
94.5
93.9
0
11.8
53.8
19.8
76.9
53.8
94.3
54
372
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
Fig. 2. Dose coverage for thoracic spine region. Lines represent medium dose coverage (95% isodose), location of maximum dose, and GTV for various techniques. (A) PA, (B) posterior wedge pair, (C) PA/AP mixed energy, and (D) IMRT.
are well within acceptable criteria regardless of the planning technique or the size of the patients. The sparing of critical structures met commonly used dose limits within our clinic when using the small patient dataset for most of the planning techniques. For the medium and large patient datasets, the posterior wedged pair had the most sparing for all structures excluding the lung. Equally weighted AP plans were generally unacceptable, faring the worst for heart, liver, normal tissues, and plan maximum doses. The unequally weighted AP plans did show improvement over the equally weighted plans but still had rather high plan maximums and heart dose maximums. The thoracic patient data seem to reflect better coverage and dose sparing when using the posterior wedge pair technique over the other conventional planning techniques. Lumbar spine The lumbar spine datasets were evaluated for overall plan maximum dose; target coverage by the 100%, 95%, and 90% isodose lines; maximum dose to the spinal cord; and volume doses to the small bowel, stomach, combined kidney, and normal tissue (Table 2c). Figure 3A–3D shows the dose distributions for single PA field, posterior wedge pair, PA/AP, and IMRT plans. As previously discussed for both the cervical and thoracic spines, the IMRT plans reflected marked improvement for the sparing of normal
tissues and critical structures over any of the conventional techniques. The coverage by the 95% isodose line was adequate with all techniques except for the single posterior field and the 6-MV posterior wedged pair. The small bowel dose increased with increasing patient separation, regardless of technique, but the combined kidney and stomach doses decreased with increasing separation. However, all were within acceptable dose tolerances as defined by QUANTEC.28 The small bowel dose was minimized most by using the posterior wedged pair technique, whereas the spinal cord and combined kidney doses were the lowest when using the parallel opposed AP fields. Discussion and Conclusion This paper demonstrates the superiority of the IMRT plans over conventional techniques for both coverage of the target volumes and the sparing of the normal tissue and critical structures. The IMRT plans resulted in a reduced volume of normal tissue receiving 67% of the prescription dose by factors ranging from 6.5–14 when compared with the conventional plans. When IMRT is not chosen as the method of treatment, the plan comparisons show that the commonly used single posterior spine port seldom covers the target volume with adequate dose as defined by ICRU 5026 and always delivers a higher dose to the spinal cord than to the vertebral body target. Maximum plan doses also often exceed 110% of the prescription dose.
Table 2(c) GTVL
Small separation: 12 cm PA 6x PA 16x PostWdgPr 6x PostWdgPr 16x PA(6x):AP(16x) 2:1 weighting IMRT 6x Medium separation: 18 cm PA 6x PA 16x PostWdgPr 6x PostWdgPr 16x PA(6x):AP(16x) 2:1 weighting IMRT 6x Large separation: 23 cm PA 6x PA 16x PostWdgPr 6x PostWdgPr 16x PA(6x):AP(16x) 2:1 weighting IMRT 6x
Spinal cord
Sm bowel
Combined kidney
Normal tissue
Max %
V95 %
Min %
Dmax %
V⬎10 % (mL)
Dmean %
Dmean %
Dmax V⬎67%
% cc
112.3 108.6 105.6 104.6 106
85.5 94.2 100 100 100
89.0 92.0 96 98.2 95.3
119.7 109.7 106.7 104.4 108.3
16.5 14.0 2.2 0 7.6
29.5 30.9 28.6 30.0 31.8
23.3 107.4 38.4 36.7 24.0
122.0 109.8 108.5 105.5 109.9
429 448 346 285 442
104.7
100
95.2
98.4
0
13.0
10.6
98.7
60
116.1 114.7 109.6 106.2 107.3
76.2 89.0 87.3 100 100
86.0 90 90.4 97.2 95.1
130.6 127.3 114.2 106.9 113.5
35.5 34.8 28.3 5.9 31.5
33.4 34.8 33.3 34.9 36.7
6.3 5.1 13.0 11.6 6.4
156.5 136.5 125.4 108.1 126.9
1353 1542 1780 1519 1661
107.2
100
95
97.0
0
13.4
7.1
98.5
132
119.2 117.3 115.6 107.1 109
75.1 86.1 845.2 100 99.7
85.9 89.0 88.8 97.1 94.2
137.0 132.9 125.9 108.2 116.3
26.4 25.8 24.1 9.0 23.5
14.4 14.9 16.0 16.7 16.4
3.9 2.6 7.4 5.9 3.7
173.4 148.8 139.2 109.4 135.4
1527 1618 2170 1973 2063
104.5
100
96
97.7
0
7.7
14.7
98.8
158
M.M. Ewing et al. / Medical Dosimetry 37 (2012) 369-373
373
Fig. 3. Dose coverage for lumber spine region. Lines represent medium dose coverage (95% isodose), location of maximum dose, and GTV for various techniques. (A) PA, (B) posterior wedge pair, (C) PA/AP mixed energy, and (D) IMRT.
For the cervical spine patients, the plan comparisons show that the anterior wedge pair meets more of the desired criteria than any of the other non-IMRT plans. The thoracic conventional plans demonstrate comparable coverage of the target by either the posterior wedge pair or parallel opposed AP with optimized weighting. Both of those planning techniques give similar doses to the spinal cord and those maximum doses ranged from 102–111% of the prescribed dose. The AP field combination gives the best sparing to the lung, but the posterior wedge pair is superior at sparing the stomach, heart, esophagus, liver, and normal tissues. The lumbar spine conventional planning techniques illustrate excellent coverage by the 95% isodose line for all but the single posterior field, and in some cases, by any plan using 6 MV rather than a higher energy. The spinal cord maximum doses are best tailored by using the AP optimized weighted beams over the posterior wedge pair, but there was very little difference among the mean doses to the spinal cord. The small bowel and stomach doses were best minimized by using the posterior wedge pair, whereas the combined kidney doses were the lowest by using the AP technique. The IMRT technique is shown to be superior to any of the other techniques but may not be the most practical for treating the metastatic patient who is in pain and cannot lie still for a long period of time. However, it does allow the doses to critical structures and normal tissues to be better controlled and it could allow the typical palliative doses to be increased, or allow “retreatment” of the same vertebral body. The technique to be used is ultimately chosen at the discretion of the treating physician, but these data do demonstrate the superiority of certain techniques based on the anatomic section of the spine to be treated and the separation of the patient. The resulting data from this exercise could be used by the treatment planner or a physician in determining which treatment technique might be better suited for their patient, based on the region of the spine involved and the size of the patient. This study could be further extended to advanced technologies and techniques, such as IMRT, SBRT, tomotherapy,10 and protons.15 We did not include the use of proton therapy (even though it is available at our center) because it is not readily available to most patients or clinics. However, it may be advantageous in many patients especially in pediatric patients.16,17,29 One should not become complacent merely because a treatment technique is considered adequate. The goal is to deliver the best treatment possible and to always strive for continuous quality improvement. References 1. Harries B. Spinal cord compression. BMJ. 1:611– 4; 1970. 2. Findlay G.F. Adverse effects of the management of malignant spinal cord compression. J. Neurol. Neurosurg. Psychiatry. 47:761– 8; 1984. 3. Harries B. Spinal cord compression. II. BMJ. 1:673– 6; 1970. 4. Windeyer B. Metastases in the central nervous system: treatment by radiotherapy and chemotherapy. Proc. Royal. Soc. Med. 57:1153–9; 1964.
5. Harel R. Angelov L. Spine metastases: current treatments and future directions. Eur. J. Cancer. 46:2696 –707; 2010. 6. Lutz S.; Berk L.; Chang E.; et al. Palliative radiotherapy for bone metastases: an ASTRO evidence-based guideline. Int. J. Radiat. Oncol. Biol. Phys. 79:965–76; 2011. 7. Chang U.K.; Youn S.M.; Park S.Q.; et al. Clinical results of cyberknife radiosurgery for spinal metastases. J. Korean. Neurosurg. Soc. 46:538 – 44; 2009. 8. Gong Y.; Wang J.; Bai S.; et al. Conventionally fractionated image-guided intensity modulated radiotherapy (IG-IMRT): a safe and effective treatment for cancer spinal metastasis. Radiol. Oncol. 11:1–10; 2008. 9. Li C.; Muller-Runkel R.; Vijayakumar S.; et al. Craniospinal axis irradiation: an improved electron technique for irradiation of the spinal axis. Br. J. Radiol. 67:186 –93; 1994. 10. McIntosh A.; Dunlap N.; Sheng K.; et al. Helical tomotherapy-based STAT RT: dosimetric evaluation for clinical implementation of a rapid radiation palliation program. Med. Dosimetry. 35:280 – 6; 2010. 11. Muller-Runkel R.; Vijayakumar, S. Spinal axis irradiation with electrons: measurements of attenuation by the spinal processes. Med. Phys. 13:539 – 43; 1986. 12. Nutting C.; Convery D.; Cosgrove V.; et al. Improvements in target coverage and reduced spinal cord irraditaion using intensity-modulated radiotherapy (IMRT) in patients with carcinoma of the thyroid gland. Radiother. Oncol. 60:173– 80; 2001. 13. Roback D.M.; Johnson J.M.; Khan F.M.; et al. The use of tertiary collimation for spinal irradiation with extended SSD electron fields. Int. J. Radiat. Oncol. Biol. Phys. 37: 1187–92; 1997. 14. Wowra B.; Muacevic A.; Zausinger S.; et al. Radiosurgery for spinal malignant tumors. Deutsches ãrzteblatt Internationaux 106:106 –12; 2009. 15. St. Clair W.H.; Adams J.A.; Bues M.; et al. Advantage of protons compared to conventional X-ray or IMRT in the treatment of a pediatric patient with medulloblastoma. Int. J. Radiat. Oncol. Biol. Phys. 58:727–34; 2004. 16. DeLaney T.F.; Liebsch N.J.; Pedlow F.X.; et al. Phase II study of high-dose photon/ proton radiotherapy in the management of spine sarcomas. Int. J. Radiat. Oncol. Biol. Phys. 74:732–9; 2009. 17. Yuh G.E.; Loredo L.N.; Yonemoto L.T.; et al. Reducing toxicity from craniospinal irradiation: using proton beams to treat medulloblastoma in young children. Cancer. J. 10:386 –90; 2004. 18. Brinkley D.; Masters, H.E. The depth of the spinal cord below the skin. Br. J. Radiol. 40:66 – 8; 1967. 19. Kubo H. Estimate of the excessive spinal cord dose in AP-PA thorax treatment. Radiother. Oncol. 13:187–92; 1989. 20. Barton R.; Robinson G.; Gutierrez E.; et al. Palliative radiation for vertebral metastases: the effect of variation in prescription parameters on the dose received at depth. Int. J. Radiat. Oncol. Biol. Phys. 52:1083–91; 2001. 21. Vijayakumar S. Treatment of metastatic spinal cord compression: an unsettled issue. J. Nat. Med. Assoc. 75:297–303; 1984. 22. Vijayakumar S.; Muller-Runkel, R. Irradiation of the thoracic esophagus. Prone versus supine treatment positions. Acta. Radiol. Oncol. 25:187–9; 1986. 23. Yin F.F.; Ryu S.; Ajlouni M.; et al. A technique of intensity-modulated radiosurgery (IMRS) for spinal tumors. Med. Phys. 29:2815–22; 2002. 24. Endicott T.; Fisher B.; Wong E.; et al. Pulmonary sequelae after electron spinal irradiation. Radiother. Oncol. 60:267–72; 2001. 25. Andic F.; Baz Cifci S.; Ors Y.; et al. A dosimetric comparison of different treatment plans of palliative spinal bone irradiation: analysis of dose coverage with respect to ICRU 50 report. J. Exp. Clin. Cancer. Res. 28:2; 2009. 26. ICRU Report 50. Prescribing, Recording, and reporting Photon Beam Therapy. Bethesda, MD: International Commission on Radiation Units and Measurements; 1993. 27. ICRU Report 83. In Prescribing, Recording, and Reporting Intensity-Modulated PhotonBeam Therapy. Bethesda, MD: International Commission on Radiation Units and Measurements; 2010. 28. Marks L.B.; Ten Haken R.K.; Martel M.K. Quantitative analyses of normal tissue effects in the clinic (QUANTEC). Int. J. Radiat. Oncol. Biol. Phys. 76;(suppl):S1–S160; 2010. 29. Athar B.S.; Paganetti, H. Comparison of second cancer risk due to out-of-field doses from 6-MV IMRT and proton therapy based on 6 pediatric patient treatment plans. Radiother. Oncol. 98:87–92; 2011.