- Email: [email protected]
A comparison of two immobilization systems for stereotactic body radiation therapy of lung tumors
Radiotherapy and Oncology 95 (2010) 103–108
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
SBRT of lung tumors
A comparison of two immobilization systems for stereotactic body radiation therapy of lung tumors Kathy Han a, Patrick Cheung a,*, Parminder S. Basran b, Ian Poon a, Latifa Yeung c,d, Fiona Lochray a a
Department of Radiation Oncology, Odette Cancer Center, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario, Canada; b Department of Medical Physics, BC Cancer Agency—Vancouver Island Center, Victoria, British Columbia, Canada; c Department of Pediatrics, University of Toronto, Toronto, Ontario, Canada; d Department of Health Policy, Management and Evaluation, University of Toronto, Toronto, Ontario, Canada
a r t i c l e
i n f o
Article history: Received 7 September 2009 Received in revised form 8 January 2010 Accepted 31 January 2010 Available online 26 February 2010 Keywords: Stereotactic body radiation therapy Respiratory tumor motion Intrafraction tumor motion Non-small cell lung cancer Bodyfix Abdominal compression plate
a b s t r a c t Purpose: This study aims to compare the efficacy, efficiency and comfort level of two immobilization systems commonly used in lung stereotactic body radiation therapy (SBRT): the Bodyfix and the abdominal compression plate (ACP). Materials and methods: Twenty-four patients undergoing SBRT for medically inoperable stage I lung cancer or pulmonary metastases were entered on this prospective randomized study. All underwent 4DCT simulation with free breathing, the Bodyfix, and the ACP to assess respiratory tumor motion. After CT simulation, patients were randomly assigned to immobilization with either the Bodyfix or the ACP for the actual SBRT treatment. Cone beam CTs (CBCTs) were acquired before and after each treatment to assess intrafraction tumor motion. Setup time and patient comfort were recorded. Results: There were 16 upper lobe, two middle lobe and seven lower-lobe lesions. Both the Bodyfix and the ACP significantly reduced the superior–inferior (SI) and overall respiratory tumor motion compared to free breathing (4.6 and 4.0 vs 5.3 mm; 5.3 and 4.7 vs 6.1 mm, respectively, p < 0.05). The ACP further reduced the SI and overall respiratory tumor motion compared to the Bodyfix (p < 0.05). The mean overall intrafraction tumor motion was 2.3 mm with the Bodyfix and 2.0 mm with the ACP (p > 0.05). The ACP was faster to set up and rated more comfortable by patients than the Bodyfix (p < 0.05). Conclusions: While there is no significant difference between the Bodyfix and ACP in reducing intrafraction tumor motion, the ACP is more comfortable, faster to set up, and superior to the Bodyfix in reducing SI and overall respiratory tumor motion. Ó 2010 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 95 (2010) 103–108
Lung cancer is the leading cause of cancer-related deaths worldwide [1]. The majority (75–85%) of patients with lung cancer have non-small cell lung cancer (NSCLC), and 15–20% of NSCLC patients have stage I/II disease at presentation. Surgery remains the treatment of choice for patients with stage I NSCLC, with crude local recurrence rate of approximately 7% [2], and 5 year survival rates of 67–82% for pathologic Stage IA (T1N0M0) NSCLC, and 57– 68% for pathologic stage IB (T2N0M0) NSCLC [2–4]. Some patients with potentially curable stage I NSCLC are unable to undergo surgery because of their compromised pulmonary reserve, cardiac function, or significant co-morbidities. For such medically inoperable patients and patients who refuse surgery, conventional radiotherapy (RT) had traditionally been offered (total dose of 50–70 Gy in 1.8–2.5 Gy daily fractions). However, the results have been disappointing. A review of 18 studies using conventional RT for stage I NSCLC revealed mean 5-year overall and
* Corresponding author. Address: Department of Radiation Oncology, Odette Cancer Center, T2-105, Sunnybrook Health Sciences Center, 2075 Bayview Avenue, Toronto, ON, Canada M4N 3M5. E-mail address: [email protected] (P. Cheung). 0167-8140/$ - see front matter Ó 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2010.01.025
cause-specific survival rates of 21 ± 8% and 25 ± 9%, respectively [5]. Local recurrence was the most common reason for treatment failure, ranging from 6.4–70% (median 40%) [5]. Local control increased with smaller tumor size and higher RT dose [5–8], and long-term survival was linked to the achievement of local control. Stereotactic body radiation therapy (SBRT) uses high dose per fraction, multiple beam angles to achieve sharp dose gradients, high-precision target localization, and techniques accounting for tumor motion to allow precise targeting and radiotherapy delivery to extracranial tumors. It enables the delivery of higher biologic effective dose (BED) over a shorter period of time than conventional RT, while minimizing the normal tissue exposure to high dose radiation. SBRT has achieved results that are much better than conventional RT, and comparable to surgical resection for stage I NSCLC. Three-year local control rate with SBRT for stage I NSCLC exceeds 85% [9]. Although results with SBRT are promising, the dose-fractionation, radiotherapy techniques and immobilization method employed across studies are highly variable. Many studies just used a vacuum cushion (either by itself or inside a stereotactic body frame [SBF]), and some applied abdominal compression for tumors with larger respiratory movement.
104
Bodyfix vs abdominal compression plate
Motion due to respiration adds complexity to SBRT for lung tumors. Whereas increasing the planning target volume margin to accommodate for respiratory motion at the expense of irradiating a larger volume of normal lung tissue is acceptable for conventional RT, it is undesirable when using high doses common to SBRT. A number of techniques have been developed to account for tumor motion, either by minimizing respiratory motion via immobilization (e.g. abdominal compression, breath hold techniques), or by accounting for physiologic tumor motion via tracking or gating the beam-on time to a particular phase of the respiratory cycle. Except for abdominal compression, most of these methods significantly increase the duration of the already long treatment time (30–45 min) for these patients who are often frail and elderly. Abdominal compression works by limiting diaphragmatic excursion, and thereby the breathing-induced motion of the tumor. It has been one of the most widely used methods of decreasing respiration-associated tumor movement in SBRT treatment. Two common immobilization devices that employ abdominal compression are the Bodyfix system, and the abdominal compression plate (ACP) (Fig. 1). A few studies have shown that the Bodyfix system offers high repositioning accuracy [10,11], and reduces respiratory tumor movement in superior–inferior (SI) and medial–lateral (ML) directions compared with free-breathing condition [12]. The ACP has also been shown to control both SI and overall motion of lung tumors when compared to no compression [13,14]. However, there has not been a direct comparison of the Bodyfix system to the ACP with respect to their effectiveness in reducing respiratory and intrafraction tumor motion. This study was designed to compare the efficacy, efficiency, and comfort level of the Bodyfix system with the ACP. Patients were enrolled onto this prospective randomized study, and the following were investigated: (i) respiratory tumor motion for all patients with no abdominal compression (i.e. free breathing) compared with the Bodyfix and the ACP; (ii) intrafraction tumor motion; (iii) setup time; and (iv) patient comfort level with each immobilization system.
Materials and methods Patient eligibility Ethical approval for this study was obtained from the research ethics board of Sunnybrook Health Sciences Center. Between October 2008 and July 2009, 24 consecutive patients undergoing lung SBRT at Sunnybrook Health Sciences Center were prospectively entered into the study. Patients were eligible for the study if they had stage I NSCLC or lung oligometastases suitable for SBRT with curative intent, and were medically inoperable or refused surgery. The maximum tumor diameter had to be 65 cm. There was no restriction for enrollment relating to the location of the lesion.
4DCT simulation and assessment of respiratory tumor motion Patients underwent four-dimensional CT (4DCT) simulation with no abdominal compression (i.e. free breathing), the Bodyfix, and the ACP. For all three 4DCT scans, the BlueBAG™ vacuum cushion that came with the Bodyfix system (Medical Intelligence, Elekta, Schwabmünchen, Germany) was used to ensure that the patient remained in the same body position for all three 4DCT scans. For each patient, the vacuum cushion was first molded to the patient’s supine position, and the initial 4DCT scan was performed with no abdominal compression (i.e. free breathing). Then, the Bodyfix system was applied and a second 4DCT scan acquired. The Bodyfix system consisted of the vacuum cushion, clear cover sheet, vacuum pump and pads. Pads were placed on the patient’s chest/abdomen for abdominal compression, and a cover sheet was placed over the patient and fixed to sides of the vacuum cushion. The air between the cover sheet, patient and vacuum cushion was evacuated using the vacuum pump with the maximum pressure tolerated by the patient. When vacuum pressure was applied, the cover sheet and pads exerted downward pressure on the chest/ abdomen, immobilizing the patient, and forcing him/her into a shallow, regular breathing pattern. After the second 4DCT scan, the coversheet, vacuum pump and pads were removed, and the ACP (Medical Intelligence, Elekta, Schwabmünchen, Germany) was placed approximately 3–4 cm below the patient’s rib cage and inferior to the xiphoid process. The abdominal compression screw was tightened to the maximum pressure tolerated by patient, and the final 4DCT scan was acquired with the ACP. 4DCT images were acquired using a Philips Brilliance wide bore system, and respiratory cycle signal were monitored by the bellows system (Philips Medical System, Wisconsin, USA). The slice thickness was 3 mm, consistent with other studies [12,15,16]. Pixel size in the transverse direction was 1.0 mm. The 4DCT image acquisitions were reconstructed in 10 equally spaced phase bins. For each patient, the 4DCT-derived peak inspiratory (CT0%), peak expiratory (CT50%), maximum intensity projection (CTMIP), and average (CTAVG) CT images were transferred to the Philips Pinnacle 8.0 treatment-planning system. Internal target volume (ITV) was defined by the fusion of the gross tumor volume (GTV) contoured on the CTMIP, CT0% and CT50%. Planning target volume (PTV) was derived by adding a 5 mm circumferential expansion of the ITV. Treatment planning was performed using the Philips Pinnacle 8.0 treatment-planning system, using the CTAVG as the primary dataset to perform dose calculation. To assess respiratory tumor motion, the gross tumor volume (GTV) was contoured by a single physician (KH) in a consistent manner on each CT0% and CT50% dataset for the free breathing, Bodyfix, and ACP scenarios, and the geometric center of each GTV volume was determined in Pinnacle. For each immobilization method, the respiratory tumor motion was defined as the absolute
Fig. 1. Bodyfix system (left) and the abdominal compression plate (right).
K. Han et al. / Radiotherapy and Oncology 95 (2010) 103–108
difference between the GTV0% and GTV50% geometric center coordinates in three dimensions [medial–lateral (ML), anterior–posterior (AP), superior–inferior (SI)]. The overall respiratory tumor motion (3D tumor motion vector) was calculated using the following formula: (DML2 + DAP2 + DSI2)1/2. Respiratory tumor motions measured with free breathing, the Bodyfix and the ACP were compared using the repeated-measures analysis of variance (ANOVA) test. The Tukey post hoc test was performed when the p value from the ANOVA test was <0.05. At the end of simulation, patients were asked to rate their comfort level with the Bodyfix and ACP on a visual scale from 1 to 10 and to indicate their preference (see Appendix 1). Comfort rating scores for the Bodyfix and ACP were compared using the paired t test. The percentage of patients with preference for each immobilization method was calculated. Radiation treatment and intrafraction tumor motion Patients received 48–60 Gy in four fractions to the tumor over an 11-day period on an Elekta Synergy linear accelerator. Standard plan configuration consisted of seven equiangular co-planar beams, plus additional non-co-planar beams when warranted. The CTAVG dataset along with the ITV and PTV contours were transferred to the treatment unit (Synergy accelerators using the XVI software, Elekta Oncology Systems, Crawley, UK) to facilitate the image guidance process. Patients were randomly assigned immobilization with either the Bodyfix or the ACP for their actual SBRT treatment. Randomization was carried out using the sealed envelope method to eliminate selection bias. All patients were treated as allocated during randomization and immobilized with the same applied abdominal pressure as during CT simulation. Prior to each fraction, a localization cone beam CT (CBCT) was acquired. The CBCT acquisition time was 60 s for one full gantry rotation, ensuring that an averaged position of the tumor was obtained over many breathing cycles. The distance between CBCT axial slices was 1.0 mm and the pixel size was 1.0 mm in the transverse direction. With a scanning energy of 120 kV and a dose of approximately 2 cGy per acquisition, the CBCT images had adequate image contrast for visualizing soft tissue tumors in the lung. To calculate the necessary shifts for image guidance, the initial CBCT was fused to the planning CT (CTAVG) using a clip box placed around the spine to register the CBCT quickly to bony anatomy. The therapists then manually adjusted the registration to ensure that the tumor on CBCT localizes well within the ITV defined on the planning CT. Each registration was visualized and approved by the attending physician (PC or IP) before starting treatment, and any necessary couch shifts or rotations were carried out remotely at the control station of the treatment unit. To assess intrafraction tumor motion, additional CBCTs were acquired. A ‘‘pre-treatment” CBCT was performed before beam-on time (after any necessary couch shifts/rotation from the initial image guidance), and a ‘‘post-treatment” CBCT was performed near the end of each treatment, (after delivery of the co-planar beams but before the couch rotation for non-co-planar beams). The preand post-treatment CBCT images for all four SBRT fractions were exported to the Philips Pinnacle 8.0 treatment-planning system . Each pre- and post-treatment CBCT dataset was fused to each other to ensure that the treatment isocenters lined up. For each patient, the tumor on each set of pre- and post-treatment CBCT images were contoured by a single physician (KH) in a consistent manner, and the geometric center of the tumor position was determined in Pinnacle. One hundred and sixty-eight CBCTs were available for analysis. Intrafraction motion was defined as the absolute difference between the post-treatment and pre-treatment geometric center coordinates in three dimensions (ML, AP, and SI). The overall
105
intrafraction motion (3D tumor motion vector) was calculated using the following formula: (DML2 + DAP2 + DSI2)1/2. Intrafraction tumor motions measured from the Bodyfix and the ACP were compared using the paired t test. Setup time (total time in treatment room minus beam-on time) for each fraction was recorded. The mean setup time for the two immobilization methods was compared using the paired t test. Results Patient and tumor characteristics Twenty-four patients with stage I NSCLC or oligometastases in the lung were enrolled onto the study. Median age was 75 (range 54–89). There were 13 male and 11 female patients. Twenty patients had medically inoperable disease due to poor lung function and/or co-morbidities and four refused surgery. One patient had two lung lesions. The distribution of tumor location was: three left upper lobe, two left lower lobe, 13 right upper lobe, two right middle lobe and five right lower lobe (Fig. 2). Median tumor size was 3 cm (range 1.5–5 cm) and median GTV from CT0% was 10.3 cc (range 1.4–90.9 cc). Respiratory tumor motion All 24 patients underwent 4DCT simulation with no abdominal compression (i.e. free breathing), the Bodyfix and the ACP. The mean (and range) of vacuum pressure applied for the Bodyfix was 94 m Bar (60–100 m Bar). The mean (and range) of screw reading for the ACP was 113 cm (45–191 cm). Fig. 3 shows the mean respiratory tumor motion determined from 4DCT with free breathing, the Bodyfix and the ACP. The mean (and range of) respiratory tumor motion for all patients with free breathing was 1 mm (0–3.4 mm) in the ML direction, 1.7 mm (0– 4.6 mm) in the AP direction, 5.3 mm (0.2–22.7 mm) in the SI direction, and 6.1 mm (0.4–23.0 mm) overall. Under free breathing, 9 out of 25 lesions (36%) moved greater than 5 mm in any one direction, and 4 out of 25 lesions (16%) moved greater than 10 mm in any one direction. The mean (and range of) respiratory tumor motion for all patients with the Bodyfix was 0.8 mm (0–3.0 mm) in the ML direction, 1.6 mm (0–6.1 mm) in the AP direction, 4.6 mm (0–22.4 mm) in the SI direction, and 5.3 mm (0–23.3 mm) overall. In one lesion, tumor motion paradoxically increased with Bodyfix from 8.7 to 11.1 mm in the SI direction when compared with free
Fig. 2. Locations of lung tumors. Numbers represent case numbers.
106
Bodyfix vs abdominal compression plate
2.7 mm) in the AP direction, 1.5 mm (0–5.7 mm) in the SI direction, and 2.3 mm (0–6.3 mm) overall. The mean intrafraction tumor motion for patients immobilized with the ACP was 0.8 mm (0– 3.9 mm) in the ML direction, 1.0 mm (0–3.2 mm) in the AP direction, 1.2 mm (0–4.6 mm) in the SI direction, and 2.0 mm (0– 5.0 mm) overall. The differences were not statistically significant using the t tests (p > 0.05). Setup time The mean setup time required for the Bodyfix and the ACP was 40 ± 21 min and 33 ± 10 min, respectively. The ACP was significantly faster to set up than the Bodyfix (p = 0.04). Fig. 3. Respiratory tumor motion with free breathing, the Bodyfix and the abdominal compression plate. Error bars represent one standard deviation.
breathing. The mean (and range of) respiratory tumor motion for all patients with the ACP was 0.8 mm (0.1–2.9 mm) in the ML direction, 1.6 mm (0.1–5.6 mm) in the AP direction, 4 mm (0– 21.5 mm) in the SI direction, and 4.7 mm (0.4–21.7 mm) overall. Repeated-measures ANOVA test showed that the respiratory tumor motion in the SI direction differed significantly among free breathing, the Bodyfix and the ACP (p < 0.0001). Tukey post hoc comparisons of the three groups indicated both the Bodyfix and the ACP significantly reduced the SI respiratory tumor motion compared to free breathing (p < 0.05 and p < 0.01, respectively), and the ACP significantly reduced the SI respiratory tumor motion compared to the Bodyfix (p < 0.05). Overall respiratory tumor motion also differed significantly among free breathing, the Bodyfix and the ACP (p < 0.0001). Both the Bodyfix and the ACP significantly reduced the overall respiratory tumor motion compared to free breathing (both p < 0.01), and the ACP significantly reduced the overall respiratory tumor motion compared to the Bodyfix (p < 0.05). There was no significant difference among the three groups in ML or AP respiratory tumor motion (p > 0.05). Intrafraction tumor motion All 24 patients underwent four fractions of treatment as allocated during randomization and pre- and post-treatment CBCTs were acquired. Half the patients were randomized to immobilization with the Bodyfix, and the other half with the ACP. Baseline patient and tumor characteristics were evenly distributed between the two groups (p > 0.05). Fig. 4 shows the mean intrafraction tumor motion measured from CBCTs with the Bodyfix and the ACP. The mean (and range of) intrafraction tumor motion for patients immobilized with the Bodyfix was 0.8 mm (0–2.5 mm) in the ML direction, 0.9 mm (0–
Fig. 4. Intrafraction tumor motion with the Bodyfix and the abdominal compression plate. Error bars represent one standard deviation.
Patient comfort Patients rated the ACP to be more comfortable than the Bodyfix (mean score of 3.2 vs 4.6, p = 0.005). The majority of patients (63%) preferred the ACP, 21% preferred the Bodyfix and 17% had no preference. Discussion This study is the first to compare the efficacy (reducing respiratory and intrafraction tumor motion), efficiency (setup time) and comfort level of two SBRT immobilization systems: the Bodyfix and the ACP. Our findings indicate that the ACP is superior to the Bodyfix in reducing SI and overall respiratory tumor motion, faster to set up, and more comfortable than the Bodyfix. There is no significant difference between the Bodyfix and ACP in reducing intrafraction tumor motion. Free breathing motion of lung tumors has been studied using various methods, including fluoroscopy, multiple portal or CT imaging over time, 4DCT and dynamic MRI. Respiratory tumor motion varies widely depending on tumor location and ranges anywhere from 0 to 28 mm [17–22]. The dominant respiratory tumor motion is usually in the SI direction, and lower lobe tumors tend to move more than upper or middle lobe tumors. A recent study using 4DCT analysis showed that for lower lung tumors, the mean ML, AP and SI movement was 1.2, 1.6, and 14.7 mm, respectively [23]. As expected, the dominant respiratory tumor motion in our study was in the SI direction (mean free breathing motion 5.3 mm), whereas lateral and AP motion were much smaller (1.0 and 1.7 mm, respectively). The majority of our patients had lesions in the upper lobe, and therefore most of the tumors moved <5 mm in the SI direction. Despite that, the Bodyfix and the ACP significantly reduced the SI and overall respiratory tumor motion compared to free breathing. Moreover, the ACP further reduced the SI and overall respiratory tumor motion compared to the Bodyfix. Although it was likely that these differences in tumor motion were due to the different immobilization devices used in our study, it was also possible that patients may have relaxed and breathed less deeply during the study procedure of performing the 4DCT scans, resulting in less breathing-induced tumor motion as patients progressed from free breathing, Bodyfix and then the ACP. It would be impossible from our study to determine whether this was the case, but we feel this was unlikely, given that the 4DCT simulation staff checked for a steady breathing trace prior to performing each 4DCT scan. Our results are consistent with that of previous studies. Baba et al. also examined the efficacy of Bodyfix on 55 lung lesions (47% in lower lung field) using fluoroscopy [12]. They found that the Bodyfix reduced the mean SI respiratory tumor motion from 9.2 to 7.5 mm, although the amplitude of reduction in the SI direction was 3 mm or more in only 27% of the patients. In our study, SI
107
K. Han et al. / Radiotherapy and Oncology 95 (2010) 103–108 Table 1 Summary of studies quantifying intrafraction tumor motion in patients receiving SBRT for lung tumors. Author (reference), institution
Pts (n)
Immobilization device
Intrafraction tumor motion Mean ± SD (maximum) mm ML
AP
SI
Overall
Grills [24], William Beaumont, USA
24
Guckenberger [25], Germany Purdie [26], Princess Margaret Hospital, Canada
24 8
Alpha-cradle* Elekta SBF* Elekta SBF* Elekta SBF
1.1 0.8 0.3 ± 1.5 (5.8) na
1.0 1.0 1.3 ± 1.9 (5.8) na
1.3 1.0 0.6 ± 1.5 (6.8) na
Present study
24
Bodyfix Abdominal compression plate
0.8 ± 0.6 (2.5) 0.8 ± 0.9 (3.9)
0.9 ± 0.8 (2.7) 1.0 ± 0.8 (3.2)
1.5 ± 1.5 (5.7) 1.2 ± 1.0 (4.6)
na na 2.8 ± 1.6 (7.2) 2.2 ± 1.2 (4.7) 5.3 ± 3.0§ (11.4) 2.3 ± 1.3 (6.3) 2.0 ± 1.2 (5.0)
Abbreviation: ML = medial–lateral; AP = anterior–posterior; SI = superior–inferior; SBF = stereotactic body frame; na = not available. Abdominal compression applied for patients with respiratory tumor motion >5 mm in any direction at simulation. Abdominal compression applied for patients with respiratory tumor motion >10 mm in any direction at simulation. Repeat CBCT imaging done within 34min. § Repeat CBCT imaging done after 34 min. *
respiratory tumor motion was reduced by 3 mm or more in only 16% of the patients. Two studies have examined the efficacy of the ACP. Heinzerling et al. compared the effectiveness of different abdominal compression levels on 10 patients with lower lobe lung and liver tumors using 4DCT [14]. Mean SI respiratory tumor motion was 12.0, 7.5, and 6.1 mm for no compression, medium compression (MC) and high compression (HC) force, respectively. Mean overall respiratory tumor motion was 13.6, 8.3, and 7.2 mm for no compression, MC, and HC, respectively. A significant difference in the control of both SI and overall tumor motion was seen with MC and HC force, and HC force further improved SI tumor motion when compared to MC. When Negoro et al. applied an abdominal plate to 10 patients with lesions that moved >5 mm in the SI direction under fluoroscopy, the mean SI respiratory tumor movement was reduced from 12.3 mm (8–20 mm) to 7.0 mm (2–11 mm) [13]. Similarly, for the nine lesions that moved >5 mm in the SI direction in our study, the ACP reduced the mean SI respiratory tumor movement from 10.9 mm (5.9–23.0 mm) to 8.6 mm (3.1– 21.5 mm). Table 1 summarizes the intrafraction tumor motion from studies that consisted of patients undergoing SBRT for lung tumors in comparison to ours. Our intrafraction tumor motion is in keeping with that from other studies [24–26]. It is interesting to point out that Purdie et al. found that the mean overall intrafraction tumor motion was 2.2 mm (SD 1.2 mm, maximum 4.7 mm) when repeat CBCT imaging was done within 34 min, and 5.3 mm (SD 3.0 mm, maximum 11.4 mm) when the interval between localization and repeat imaging exceeded 34 min (p < 0.01) [26]. This is not unexpected and highlights the importance of a good immobilization system that will minimize the intrafraction tumor motion even with prolonged treatment time. All the other studies only used abdominal compression when respiratory tumor motion was greater than 5 or 10 mm, whereas we used abdominal compression for all our patients (pads in the Bodyfix system or the ACP). It is likely that our maximum intrafraction tumor motion is smaller than that reported by Guckenberger et al. and Purdie et al. because we applied abdominal compression on all our patients [25,26]. It is recognized that the methodology (of using before and after treatment CBCTs) used in our study to measure intrafraction motion does not capture real time breathing-induced tumor motion or patient shifts during treatment. Rather, it is a measure of the ‘‘net intrafraction tumor motion” which measures whether there is a change in the position of the tumor at the end of treatment, compared to the tumor’s position prior to treatment. In addition to efficacy in reducing respiratory and intrafraction tumor motion, patient comfort is another important parameter when comparing immobilization devices. However, no previous studies have addressed patient comfort with the Bodyfix or the
ACP. Our comparison of the Bodyfix and the ACP reveals noticeable differences regarding patient comfort, with the ACP rated significantly more comfortable than the Bodyfix and the majority of patients preferring the ACP. Moreover, the ACP was significantly faster to set up than the Bodyfix. The ACP became available at our center just prior to the opening of study, and our radiation therapists quickly found the ACP easier to use than the Bodyfix, even though they had been using the Bodyfix for half a year prior to the ACP. Conclusion Although there is no significant difference between the Bodyfix and ACP in reducing intrafraction tumor motion, the ACP is superior to the Bodyfix in terms of its efficacy in reducing SI and overall respiratory tumor motion, efficiency and comfort level. As such, the ACP may be the preferred method of immobilization for patients undergoing SBRT for lung tumors. Conflict of interest statement The authors declare that conflicts of interest do not exist. Acknowledgements We thank patients who participated in this study, and Darby Erler, Ada Wong, Naila Devji, Derek Hyde and radiation therapists at Sunnybrook Health Sciences Center for their assistance. Appendix. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.radonc.2010.01.025. References [1] Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA. Cancer J Clin 2005;55:74–108. [2] Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage I lung cancer. J Thorac Cardiovasc Surg 1995;109:120–9. [3] Naruke T, Goya T, Tsuchiya R, Suemasu K. Prognosis and survival in resected lung carcinoma based on the new international staging system. J Thorac Cardiovasc Surg 1988;96:440–7. [4] Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710–7. [5] Qiao X, Tullgren O, Lax I, Sirzén F, Lewensohn R. The role of radiotherapy in treatment of stage I non-small cell lung cancer. Lung Cancer 2003;41:1–11. [6] Dosoretz D, Katin M, Blitzer P, et al. Radiation therapy in the management of medically inoperable carcinoma of the lung: results and implications for future treatment strategies. Int J Radiat Oncol Biol Phys 1992;24:3–9.
108
Bodyfix vs abdominal compression plate
[7] Kaskowitz L, Graham M, Emami B, Halverson KJ, Rush C. Radiation therapy alone for stage I non-small cell lung cancer. Int J Radiat Oncol Biol Phys 1993;27:517–23. [8] Sibley G, Jamieson T, Marks L, Anscher MS, Prosnitz LR. Radiotherapy alone for medically inoperable stage I non-small-cell lung cancer: the Duke experience. Int J Radiat Oncol Biol Phys 1998;40:149–54. [9] Haasbeek CJ, Senan S, Smit EF, Paul MA, Slotman BJ, Lagerwaard FJ. Critical review of nonsurgical treatment options for stage I non-small cell lung cancer. Oncologist 2008;13:309–19. [10] Fuss M, Salter BJ, Rassiah P, Cheek D, Cavanaugh SX, Herman TS. Repositioning accuracy of a commercially available double-vacuum whole body immobilization system for stereotactic body radiation therapy. Technol Cancer Res Treat 2004;3:59–67. [11] Nevinny-Stickel M, Sweeney RA, Bale RJ, Posch A, Auberger T, Lukas P. Reproducibility of patient positioning for fractionated extracranial stereotactic radiotherapy using a double-vacuum technique. Strahlenther Onkol 2004;180:117–22. [12] Baba F, Shibamoto Y, Tomita N, et al. Stereotactic body radiotherapy for stage I lung cancer, small lung metastasis: evaluation of an immobilization system for suppression of respiratory tumor movement, preliminary results. Radiat Oncol 2009;4:15. [13] Negoro Y, Nagata Y, Aoki T, et al. The effectiveness of an immobilization device in conformal radiotherapy for lung tumor: reduction of respiratory tumor movement and evaluation of the daily setup accuracy. Int J Radiat Oncol Biol Phys 2008;50:889–98. [14] Heinzerling J, Anderson JF, Papiez L, et al. Four-dimensional computed tomography scan analysis of tumor and organ motion at varying levels of abdominal compression during stereotactic treatment of lung and liver. Int J Radiat Oncol Biol Phys 2008;70:1571–8. [15] Underberg RWM, Lagerwaard FJ, Slotman BJ, et al. Use of maximum intensity projections (MIP) for target volume generation in 4DCT scans for lung cancer. Int J Radiat Oncol Biol Phys 2005;63:253–60.
[16] Rietzel E, Liu AK, Doppke KP, et al. Design of 4D treatment planning target volumes. Int J Radiat Oncol Biol Phys 2006;66:287–95. [17] Sixel KE, Ruschin M, Tirona R, Cheung PC. Digital fluoroscopy to quantify lung tumor motion: Potential for patient-specific planning target volumes. Int J Radiat Oncol Biol Phys 2003;57:717–23. [18] Ekberg L, Holmberg O, Wittgren L, Bjelkengren G, Landberg T. What margins should be added to the clinical target volume in radiotherapy treatment planning for lung cancer. Radiother Oncol 1998;48:71–7. [19] Erridge SC, Seppenwoolde Y, Muller S, et al. Portal imaging to assess set-up errors, tumor motion and tumor shrinkage during conformal radiotherapy of non-small cell lung cancer. Radiother Oncol 2003;66:75–85. [20] Plathow C, Ley S, Fink C, et al. Analysis of intrathoracic tumor mobility during whole breathing cycle by dynamic MRI. Int J Radiat Oncol Biol Phys 2004;59:952–9. [21] Weiss E, Wijesooriya K, Vaughn Dill S, Keall PJ. Tumor and normal tissue motion in the thorax during respiration: analysis of volumetric and positional variations using 4DCT. Int J Radiat Oncol Biol Phys 2007;67:297–307. [22] Guckenberger M, Wilbert J, Meyer J, Baier K, Richter A, Flentje M. Is a single respiratory correlated 4D-CT study sufficient for evaluation of breathing motion? Int J Radiat Oncol Biol Phys 2007;67:1352–9. [23] Liu H, Balter P, Tutt T, et al. Assessing respiration-induced tumor motion and internal target volume using four-dimensional computed tomography for radiotherapy of lung cancer. Int J Radiat Oncol Biol Phys 2007;68:531–40. [24] Grills IS, Hugo G, Kestin LL, et al. Image-guided radiotherapy via daily online cone-beam CT substantially reduced margin requirements for stereotactic lung radiotherapy. Int J Radiat Oncol Biol Phys 2008;70:1045–56. [25] Guckenberger M, Meyer J, Wilbert J, et al. Intra-fractional uncertainties in cone-beam CT based image-guided radiotherapy (IGRT) of pulmonary tumors. Radiother Oncol 2007;83:57–64. [26] Purdie TG, Bissonnette JP, Franks K, et al. Cone-beam computed tomography for on-line image guidance of lung stereotactic radiotherapy: localization, verification, and intrafraction tumor position. Int J Radiat Oncol Biol Phys 2007;68: 243–52.
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
SBRT of lung tumors
A comparison of two immobilization systems for stereotactic body radiation therapy of lung tumors Kathy Han a, Patrick Cheung a,*, Parminder S. Basran b, Ian Poon a, Latifa Yeung c,d, Fiona Lochray a a
Department of Radiation Oncology, Odette Cancer Center, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario, Canada; b Department of Medical Physics, BC Cancer Agency—Vancouver Island Center, Victoria, British Columbia, Canada; c Department of Pediatrics, University of Toronto, Toronto, Ontario, Canada; d Department of Health Policy, Management and Evaluation, University of Toronto, Toronto, Ontario, Canada
a r t i c l e
i n f o
Article history: Received 7 September 2009 Received in revised form 8 January 2010 Accepted 31 January 2010 Available online 26 February 2010 Keywords: Stereotactic body radiation therapy Respiratory tumor motion Intrafraction tumor motion Non-small cell lung cancer Bodyfix Abdominal compression plate
a b s t r a c t Purpose: This study aims to compare the efficacy, efficiency and comfort level of two immobilization systems commonly used in lung stereotactic body radiation therapy (SBRT): the Bodyfix and the abdominal compression plate (ACP). Materials and methods: Twenty-four patients undergoing SBRT for medically inoperable stage I lung cancer or pulmonary metastases were entered on this prospective randomized study. All underwent 4DCT simulation with free breathing, the Bodyfix, and the ACP to assess respiratory tumor motion. After CT simulation, patients were randomly assigned to immobilization with either the Bodyfix or the ACP for the actual SBRT treatment. Cone beam CTs (CBCTs) were acquired before and after each treatment to assess intrafraction tumor motion. Setup time and patient comfort were recorded. Results: There were 16 upper lobe, two middle lobe and seven lower-lobe lesions. Both the Bodyfix and the ACP significantly reduced the superior–inferior (SI) and overall respiratory tumor motion compared to free breathing (4.6 and 4.0 vs 5.3 mm; 5.3 and 4.7 vs 6.1 mm, respectively, p < 0.05). The ACP further reduced the SI and overall respiratory tumor motion compared to the Bodyfix (p < 0.05). The mean overall intrafraction tumor motion was 2.3 mm with the Bodyfix and 2.0 mm with the ACP (p > 0.05). The ACP was faster to set up and rated more comfortable by patients than the Bodyfix (p < 0.05). Conclusions: While there is no significant difference between the Bodyfix and ACP in reducing intrafraction tumor motion, the ACP is more comfortable, faster to set up, and superior to the Bodyfix in reducing SI and overall respiratory tumor motion. Ó 2010 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 95 (2010) 103–108
Lung cancer is the leading cause of cancer-related deaths worldwide [1]. The majority (75–85%) of patients with lung cancer have non-small cell lung cancer (NSCLC), and 15–20% of NSCLC patients have stage I/II disease at presentation. Surgery remains the treatment of choice for patients with stage I NSCLC, with crude local recurrence rate of approximately 7% [2], and 5 year survival rates of 67–82% for pathologic Stage IA (T1N0M0) NSCLC, and 57– 68% for pathologic stage IB (T2N0M0) NSCLC [2–4]. Some patients with potentially curable stage I NSCLC are unable to undergo surgery because of their compromised pulmonary reserve, cardiac function, or significant co-morbidities. For such medically inoperable patients and patients who refuse surgery, conventional radiotherapy (RT) had traditionally been offered (total dose of 50–70 Gy in 1.8–2.5 Gy daily fractions). However, the results have been disappointing. A review of 18 studies using conventional RT for stage I NSCLC revealed mean 5-year overall and
* Corresponding author. Address: Department of Radiation Oncology, Odette Cancer Center, T2-105, Sunnybrook Health Sciences Center, 2075 Bayview Avenue, Toronto, ON, Canada M4N 3M5. E-mail address: [email protected] (P. Cheung). 0167-8140/$ - see front matter Ó 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2010.01.025
cause-specific survival rates of 21 ± 8% and 25 ± 9%, respectively [5]. Local recurrence was the most common reason for treatment failure, ranging from 6.4–70% (median 40%) [5]. Local control increased with smaller tumor size and higher RT dose [5–8], and long-term survival was linked to the achievement of local control. Stereotactic body radiation therapy (SBRT) uses high dose per fraction, multiple beam angles to achieve sharp dose gradients, high-precision target localization, and techniques accounting for tumor motion to allow precise targeting and radiotherapy delivery to extracranial tumors. It enables the delivery of higher biologic effective dose (BED) over a shorter period of time than conventional RT, while minimizing the normal tissue exposure to high dose radiation. SBRT has achieved results that are much better than conventional RT, and comparable to surgical resection for stage I NSCLC. Three-year local control rate with SBRT for stage I NSCLC exceeds 85% [9]. Although results with SBRT are promising, the dose-fractionation, radiotherapy techniques and immobilization method employed across studies are highly variable. Many studies just used a vacuum cushion (either by itself or inside a stereotactic body frame [SBF]), and some applied abdominal compression for tumors with larger respiratory movement.
104
Bodyfix vs abdominal compression plate
Motion due to respiration adds complexity to SBRT for lung tumors. Whereas increasing the planning target volume margin to accommodate for respiratory motion at the expense of irradiating a larger volume of normal lung tissue is acceptable for conventional RT, it is undesirable when using high doses common to SBRT. A number of techniques have been developed to account for tumor motion, either by minimizing respiratory motion via immobilization (e.g. abdominal compression, breath hold techniques), or by accounting for physiologic tumor motion via tracking or gating the beam-on time to a particular phase of the respiratory cycle. Except for abdominal compression, most of these methods significantly increase the duration of the already long treatment time (30–45 min) for these patients who are often frail and elderly. Abdominal compression works by limiting diaphragmatic excursion, and thereby the breathing-induced motion of the tumor. It has been one of the most widely used methods of decreasing respiration-associated tumor movement in SBRT treatment. Two common immobilization devices that employ abdominal compression are the Bodyfix system, and the abdominal compression plate (ACP) (Fig. 1). A few studies have shown that the Bodyfix system offers high repositioning accuracy [10,11], and reduces respiratory tumor movement in superior–inferior (SI) and medial–lateral (ML) directions compared with free-breathing condition [12]. The ACP has also been shown to control both SI and overall motion of lung tumors when compared to no compression [13,14]. However, there has not been a direct comparison of the Bodyfix system to the ACP with respect to their effectiveness in reducing respiratory and intrafraction tumor motion. This study was designed to compare the efficacy, efficiency, and comfort level of the Bodyfix system with the ACP. Patients were enrolled onto this prospective randomized study, and the following were investigated: (i) respiratory tumor motion for all patients with no abdominal compression (i.e. free breathing) compared with the Bodyfix and the ACP; (ii) intrafraction tumor motion; (iii) setup time; and (iv) patient comfort level with each immobilization system.
Materials and methods Patient eligibility Ethical approval for this study was obtained from the research ethics board of Sunnybrook Health Sciences Center. Between October 2008 and July 2009, 24 consecutive patients undergoing lung SBRT at Sunnybrook Health Sciences Center were prospectively entered into the study. Patients were eligible for the study if they had stage I NSCLC or lung oligometastases suitable for SBRT with curative intent, and were medically inoperable or refused surgery. The maximum tumor diameter had to be 65 cm. There was no restriction for enrollment relating to the location of the lesion.
4DCT simulation and assessment of respiratory tumor motion Patients underwent four-dimensional CT (4DCT) simulation with no abdominal compression (i.e. free breathing), the Bodyfix, and the ACP. For all three 4DCT scans, the BlueBAG™ vacuum cushion that came with the Bodyfix system (Medical Intelligence, Elekta, Schwabmünchen, Germany) was used to ensure that the patient remained in the same body position for all three 4DCT scans. For each patient, the vacuum cushion was first molded to the patient’s supine position, and the initial 4DCT scan was performed with no abdominal compression (i.e. free breathing). Then, the Bodyfix system was applied and a second 4DCT scan acquired. The Bodyfix system consisted of the vacuum cushion, clear cover sheet, vacuum pump and pads. Pads were placed on the patient’s chest/abdomen for abdominal compression, and a cover sheet was placed over the patient and fixed to sides of the vacuum cushion. The air between the cover sheet, patient and vacuum cushion was evacuated using the vacuum pump with the maximum pressure tolerated by the patient. When vacuum pressure was applied, the cover sheet and pads exerted downward pressure on the chest/ abdomen, immobilizing the patient, and forcing him/her into a shallow, regular breathing pattern. After the second 4DCT scan, the coversheet, vacuum pump and pads were removed, and the ACP (Medical Intelligence, Elekta, Schwabmünchen, Germany) was placed approximately 3–4 cm below the patient’s rib cage and inferior to the xiphoid process. The abdominal compression screw was tightened to the maximum pressure tolerated by patient, and the final 4DCT scan was acquired with the ACP. 4DCT images were acquired using a Philips Brilliance wide bore system, and respiratory cycle signal were monitored by the bellows system (Philips Medical System, Wisconsin, USA). The slice thickness was 3 mm, consistent with other studies [12,15,16]. Pixel size in the transverse direction was 1.0 mm. The 4DCT image acquisitions were reconstructed in 10 equally spaced phase bins. For each patient, the 4DCT-derived peak inspiratory (CT0%), peak expiratory (CT50%), maximum intensity projection (CTMIP), and average (CTAVG) CT images were transferred to the Philips Pinnacle 8.0 treatment-planning system. Internal target volume (ITV) was defined by the fusion of the gross tumor volume (GTV) contoured on the CTMIP, CT0% and CT50%. Planning target volume (PTV) was derived by adding a 5 mm circumferential expansion of the ITV. Treatment planning was performed using the Philips Pinnacle 8.0 treatment-planning system, using the CTAVG as the primary dataset to perform dose calculation. To assess respiratory tumor motion, the gross tumor volume (GTV) was contoured by a single physician (KH) in a consistent manner on each CT0% and CT50% dataset for the free breathing, Bodyfix, and ACP scenarios, and the geometric center of each GTV volume was determined in Pinnacle. For each immobilization method, the respiratory tumor motion was defined as the absolute
Fig. 1. Bodyfix system (left) and the abdominal compression plate (right).
K. Han et al. / Radiotherapy and Oncology 95 (2010) 103–108
difference between the GTV0% and GTV50% geometric center coordinates in three dimensions [medial–lateral (ML), anterior–posterior (AP), superior–inferior (SI)]. The overall respiratory tumor motion (3D tumor motion vector) was calculated using the following formula: (DML2 + DAP2 + DSI2)1/2. Respiratory tumor motions measured with free breathing, the Bodyfix and the ACP were compared using the repeated-measures analysis of variance (ANOVA) test. The Tukey post hoc test was performed when the p value from the ANOVA test was <0.05. At the end of simulation, patients were asked to rate their comfort level with the Bodyfix and ACP on a visual scale from 1 to 10 and to indicate their preference (see Appendix 1). Comfort rating scores for the Bodyfix and ACP were compared using the paired t test. The percentage of patients with preference for each immobilization method was calculated. Radiation treatment and intrafraction tumor motion Patients received 48–60 Gy in four fractions to the tumor over an 11-day period on an Elekta Synergy linear accelerator. Standard plan configuration consisted of seven equiangular co-planar beams, plus additional non-co-planar beams when warranted. The CTAVG dataset along with the ITV and PTV contours were transferred to the treatment unit (Synergy accelerators using the XVI software, Elekta Oncology Systems, Crawley, UK) to facilitate the image guidance process. Patients were randomly assigned immobilization with either the Bodyfix or the ACP for their actual SBRT treatment. Randomization was carried out using the sealed envelope method to eliminate selection bias. All patients were treated as allocated during randomization and immobilized with the same applied abdominal pressure as during CT simulation. Prior to each fraction, a localization cone beam CT (CBCT) was acquired. The CBCT acquisition time was 60 s for one full gantry rotation, ensuring that an averaged position of the tumor was obtained over many breathing cycles. The distance between CBCT axial slices was 1.0 mm and the pixel size was 1.0 mm in the transverse direction. With a scanning energy of 120 kV and a dose of approximately 2 cGy per acquisition, the CBCT images had adequate image contrast for visualizing soft tissue tumors in the lung. To calculate the necessary shifts for image guidance, the initial CBCT was fused to the planning CT (CTAVG) using a clip box placed around the spine to register the CBCT quickly to bony anatomy. The therapists then manually adjusted the registration to ensure that the tumor on CBCT localizes well within the ITV defined on the planning CT. Each registration was visualized and approved by the attending physician (PC or IP) before starting treatment, and any necessary couch shifts or rotations were carried out remotely at the control station of the treatment unit. To assess intrafraction tumor motion, additional CBCTs were acquired. A ‘‘pre-treatment” CBCT was performed before beam-on time (after any necessary couch shifts/rotation from the initial image guidance), and a ‘‘post-treatment” CBCT was performed near the end of each treatment, (after delivery of the co-planar beams but before the couch rotation for non-co-planar beams). The preand post-treatment CBCT images for all four SBRT fractions were exported to the Philips Pinnacle 8.0 treatment-planning system . Each pre- and post-treatment CBCT dataset was fused to each other to ensure that the treatment isocenters lined up. For each patient, the tumor on each set of pre- and post-treatment CBCT images were contoured by a single physician (KH) in a consistent manner, and the geometric center of the tumor position was determined in Pinnacle. One hundred and sixty-eight CBCTs were available for analysis. Intrafraction motion was defined as the absolute difference between the post-treatment and pre-treatment geometric center coordinates in three dimensions (ML, AP, and SI). The overall
105
intrafraction motion (3D tumor motion vector) was calculated using the following formula: (DML2 + DAP2 + DSI2)1/2. Intrafraction tumor motions measured from the Bodyfix and the ACP were compared using the paired t test. Setup time (total time in treatment room minus beam-on time) for each fraction was recorded. The mean setup time for the two immobilization methods was compared using the paired t test. Results Patient and tumor characteristics Twenty-four patients with stage I NSCLC or oligometastases in the lung were enrolled onto the study. Median age was 75 (range 54–89). There were 13 male and 11 female patients. Twenty patients had medically inoperable disease due to poor lung function and/or co-morbidities and four refused surgery. One patient had two lung lesions. The distribution of tumor location was: three left upper lobe, two left lower lobe, 13 right upper lobe, two right middle lobe and five right lower lobe (Fig. 2). Median tumor size was 3 cm (range 1.5–5 cm) and median GTV from CT0% was 10.3 cc (range 1.4–90.9 cc). Respiratory tumor motion All 24 patients underwent 4DCT simulation with no abdominal compression (i.e. free breathing), the Bodyfix and the ACP. The mean (and range) of vacuum pressure applied for the Bodyfix was 94 m Bar (60–100 m Bar). The mean (and range) of screw reading for the ACP was 113 cm (45–191 cm). Fig. 3 shows the mean respiratory tumor motion determined from 4DCT with free breathing, the Bodyfix and the ACP. The mean (and range of) respiratory tumor motion for all patients with free breathing was 1 mm (0–3.4 mm) in the ML direction, 1.7 mm (0– 4.6 mm) in the AP direction, 5.3 mm (0.2–22.7 mm) in the SI direction, and 6.1 mm (0.4–23.0 mm) overall. Under free breathing, 9 out of 25 lesions (36%) moved greater than 5 mm in any one direction, and 4 out of 25 lesions (16%) moved greater than 10 mm in any one direction. The mean (and range of) respiratory tumor motion for all patients with the Bodyfix was 0.8 mm (0–3.0 mm) in the ML direction, 1.6 mm (0–6.1 mm) in the AP direction, 4.6 mm (0–22.4 mm) in the SI direction, and 5.3 mm (0–23.3 mm) overall. In one lesion, tumor motion paradoxically increased with Bodyfix from 8.7 to 11.1 mm in the SI direction when compared with free
Fig. 2. Locations of lung tumors. Numbers represent case numbers.
106
Bodyfix vs abdominal compression plate
2.7 mm) in the AP direction, 1.5 mm (0–5.7 mm) in the SI direction, and 2.3 mm (0–6.3 mm) overall. The mean intrafraction tumor motion for patients immobilized with the ACP was 0.8 mm (0– 3.9 mm) in the ML direction, 1.0 mm (0–3.2 mm) in the AP direction, 1.2 mm (0–4.6 mm) in the SI direction, and 2.0 mm (0– 5.0 mm) overall. The differences were not statistically significant using the t tests (p > 0.05). Setup time The mean setup time required for the Bodyfix and the ACP was 40 ± 21 min and 33 ± 10 min, respectively. The ACP was significantly faster to set up than the Bodyfix (p = 0.04). Fig. 3. Respiratory tumor motion with free breathing, the Bodyfix and the abdominal compression plate. Error bars represent one standard deviation.
breathing. The mean (and range of) respiratory tumor motion for all patients with the ACP was 0.8 mm (0.1–2.9 mm) in the ML direction, 1.6 mm (0.1–5.6 mm) in the AP direction, 4 mm (0– 21.5 mm) in the SI direction, and 4.7 mm (0.4–21.7 mm) overall. Repeated-measures ANOVA test showed that the respiratory tumor motion in the SI direction differed significantly among free breathing, the Bodyfix and the ACP (p < 0.0001). Tukey post hoc comparisons of the three groups indicated both the Bodyfix and the ACP significantly reduced the SI respiratory tumor motion compared to free breathing (p < 0.05 and p < 0.01, respectively), and the ACP significantly reduced the SI respiratory tumor motion compared to the Bodyfix (p < 0.05). Overall respiratory tumor motion also differed significantly among free breathing, the Bodyfix and the ACP (p < 0.0001). Both the Bodyfix and the ACP significantly reduced the overall respiratory tumor motion compared to free breathing (both p < 0.01), and the ACP significantly reduced the overall respiratory tumor motion compared to the Bodyfix (p < 0.05). There was no significant difference among the three groups in ML or AP respiratory tumor motion (p > 0.05). Intrafraction tumor motion All 24 patients underwent four fractions of treatment as allocated during randomization and pre- and post-treatment CBCTs were acquired. Half the patients were randomized to immobilization with the Bodyfix, and the other half with the ACP. Baseline patient and tumor characteristics were evenly distributed between the two groups (p > 0.05). Fig. 4 shows the mean intrafraction tumor motion measured from CBCTs with the Bodyfix and the ACP. The mean (and range of) intrafraction tumor motion for patients immobilized with the Bodyfix was 0.8 mm (0–2.5 mm) in the ML direction, 0.9 mm (0–
Fig. 4. Intrafraction tumor motion with the Bodyfix and the abdominal compression plate. Error bars represent one standard deviation.
Patient comfort Patients rated the ACP to be more comfortable than the Bodyfix (mean score of 3.2 vs 4.6, p = 0.005). The majority of patients (63%) preferred the ACP, 21% preferred the Bodyfix and 17% had no preference. Discussion This study is the first to compare the efficacy (reducing respiratory and intrafraction tumor motion), efficiency (setup time) and comfort level of two SBRT immobilization systems: the Bodyfix and the ACP. Our findings indicate that the ACP is superior to the Bodyfix in reducing SI and overall respiratory tumor motion, faster to set up, and more comfortable than the Bodyfix. There is no significant difference between the Bodyfix and ACP in reducing intrafraction tumor motion. Free breathing motion of lung tumors has been studied using various methods, including fluoroscopy, multiple portal or CT imaging over time, 4DCT and dynamic MRI. Respiratory tumor motion varies widely depending on tumor location and ranges anywhere from 0 to 28 mm [17–22]. The dominant respiratory tumor motion is usually in the SI direction, and lower lobe tumors tend to move more than upper or middle lobe tumors. A recent study using 4DCT analysis showed that for lower lung tumors, the mean ML, AP and SI movement was 1.2, 1.6, and 14.7 mm, respectively [23]. As expected, the dominant respiratory tumor motion in our study was in the SI direction (mean free breathing motion 5.3 mm), whereas lateral and AP motion were much smaller (1.0 and 1.7 mm, respectively). The majority of our patients had lesions in the upper lobe, and therefore most of the tumors moved <5 mm in the SI direction. Despite that, the Bodyfix and the ACP significantly reduced the SI and overall respiratory tumor motion compared to free breathing. Moreover, the ACP further reduced the SI and overall respiratory tumor motion compared to the Bodyfix. Although it was likely that these differences in tumor motion were due to the different immobilization devices used in our study, it was also possible that patients may have relaxed and breathed less deeply during the study procedure of performing the 4DCT scans, resulting in less breathing-induced tumor motion as patients progressed from free breathing, Bodyfix and then the ACP. It would be impossible from our study to determine whether this was the case, but we feel this was unlikely, given that the 4DCT simulation staff checked for a steady breathing trace prior to performing each 4DCT scan. Our results are consistent with that of previous studies. Baba et al. also examined the efficacy of Bodyfix on 55 lung lesions (47% in lower lung field) using fluoroscopy [12]. They found that the Bodyfix reduced the mean SI respiratory tumor motion from 9.2 to 7.5 mm, although the amplitude of reduction in the SI direction was 3 mm or more in only 27% of the patients. In our study, SI
107
K. Han et al. / Radiotherapy and Oncology 95 (2010) 103–108 Table 1 Summary of studies quantifying intrafraction tumor motion in patients receiving SBRT for lung tumors. Author (reference), institution
Pts (n)
Immobilization device
Intrafraction tumor motion Mean ± SD (maximum) mm ML
AP
SI
Overall
Grills [24], William Beaumont, USA
24
Guckenberger [25], Germany Purdie [26], Princess Margaret Hospital, Canada
24 8
Alpha-cradle* Elekta SBF* Elekta SBF* Elekta SBF
1.1 0.8 0.3 ± 1.5 (5.8) na
1.0 1.0 1.3 ± 1.9 (5.8) na
1.3 1.0 0.6 ± 1.5 (6.8) na
Present study
24
Bodyfix Abdominal compression plate
0.8 ± 0.6 (2.5) 0.8 ± 0.9 (3.9)
0.9 ± 0.8 (2.7) 1.0 ± 0.8 (3.2)
1.5 ± 1.5 (5.7) 1.2 ± 1.0 (4.6)
na na 2.8 ± 1.6 (7.2) 2.2 ± 1.2 (4.7) 5.3 ± 3.0§ (11.4) 2.3 ± 1.3 (6.3) 2.0 ± 1.2 (5.0)
Abbreviation: ML = medial–lateral; AP = anterior–posterior; SI = superior–inferior; SBF = stereotactic body frame; na = not available. Abdominal compression applied for patients with respiratory tumor motion >5 mm in any direction at simulation. Abdominal compression applied for patients with respiratory tumor motion >10 mm in any direction at simulation. Repeat CBCT imaging done within 34min. § Repeat CBCT imaging done after 34 min. *
respiratory tumor motion was reduced by 3 mm or more in only 16% of the patients. Two studies have examined the efficacy of the ACP. Heinzerling et al. compared the effectiveness of different abdominal compression levels on 10 patients with lower lobe lung and liver tumors using 4DCT [14]. Mean SI respiratory tumor motion was 12.0, 7.5, and 6.1 mm for no compression, medium compression (MC) and high compression (HC) force, respectively. Mean overall respiratory tumor motion was 13.6, 8.3, and 7.2 mm for no compression, MC, and HC, respectively. A significant difference in the control of both SI and overall tumor motion was seen with MC and HC force, and HC force further improved SI tumor motion when compared to MC. When Negoro et al. applied an abdominal plate to 10 patients with lesions that moved >5 mm in the SI direction under fluoroscopy, the mean SI respiratory tumor movement was reduced from 12.3 mm (8–20 mm) to 7.0 mm (2–11 mm) [13]. Similarly, for the nine lesions that moved >5 mm in the SI direction in our study, the ACP reduced the mean SI respiratory tumor movement from 10.9 mm (5.9–23.0 mm) to 8.6 mm (3.1– 21.5 mm). Table 1 summarizes the intrafraction tumor motion from studies that consisted of patients undergoing SBRT for lung tumors in comparison to ours. Our intrafraction tumor motion is in keeping with that from other studies [24–26]. It is interesting to point out that Purdie et al. found that the mean overall intrafraction tumor motion was 2.2 mm (SD 1.2 mm, maximum 4.7 mm) when repeat CBCT imaging was done within 34 min, and 5.3 mm (SD 3.0 mm, maximum 11.4 mm) when the interval between localization and repeat imaging exceeded 34 min (p < 0.01) [26]. This is not unexpected and highlights the importance of a good immobilization system that will minimize the intrafraction tumor motion even with prolonged treatment time. All the other studies only used abdominal compression when respiratory tumor motion was greater than 5 or 10 mm, whereas we used abdominal compression for all our patients (pads in the Bodyfix system or the ACP). It is likely that our maximum intrafraction tumor motion is smaller than that reported by Guckenberger et al. and Purdie et al. because we applied abdominal compression on all our patients [25,26]. It is recognized that the methodology (of using before and after treatment CBCTs) used in our study to measure intrafraction motion does not capture real time breathing-induced tumor motion or patient shifts during treatment. Rather, it is a measure of the ‘‘net intrafraction tumor motion” which measures whether there is a change in the position of the tumor at the end of treatment, compared to the tumor’s position prior to treatment. In addition to efficacy in reducing respiratory and intrafraction tumor motion, patient comfort is another important parameter when comparing immobilization devices. However, no previous studies have addressed patient comfort with the Bodyfix or the
ACP. Our comparison of the Bodyfix and the ACP reveals noticeable differences regarding patient comfort, with the ACP rated significantly more comfortable than the Bodyfix and the majority of patients preferring the ACP. Moreover, the ACP was significantly faster to set up than the Bodyfix. The ACP became available at our center just prior to the opening of study, and our radiation therapists quickly found the ACP easier to use than the Bodyfix, even though they had been using the Bodyfix for half a year prior to the ACP. Conclusion Although there is no significant difference between the Bodyfix and ACP in reducing intrafraction tumor motion, the ACP is superior to the Bodyfix in terms of its efficacy in reducing SI and overall respiratory tumor motion, efficiency and comfort level. As such, the ACP may be the preferred method of immobilization for patients undergoing SBRT for lung tumors. Conflict of interest statement The authors declare that conflicts of interest do not exist. Acknowledgements We thank patients who participated in this study, and Darby Erler, Ada Wong, Naila Devji, Derek Hyde and radiation therapists at Sunnybrook Health Sciences Center for their assistance. Appendix. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.radonc.2010.01.025. References [1] Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA. Cancer J Clin 2005;55:74–108. [2] Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage I lung cancer. J Thorac Cardiovasc Surg 1995;109:120–9. [3] Naruke T, Goya T, Tsuchiya R, Suemasu K. Prognosis and survival in resected lung carcinoma based on the new international staging system. J Thorac Cardiovasc Surg 1988;96:440–7. [4] Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710–7. [5] Qiao X, Tullgren O, Lax I, Sirzén F, Lewensohn R. The role of radiotherapy in treatment of stage I non-small cell lung cancer. Lung Cancer 2003;41:1–11. [6] Dosoretz D, Katin M, Blitzer P, et al. Radiation therapy in the management of medically inoperable carcinoma of the lung: results and implications for future treatment strategies. Int J Radiat Oncol Biol Phys 1992;24:3–9.
108
Bodyfix vs abdominal compression plate
[7] Kaskowitz L, Graham M, Emami B, Halverson KJ, Rush C. Radiation therapy alone for stage I non-small cell lung cancer. Int J Radiat Oncol Biol Phys 1993;27:517–23. [8] Sibley G, Jamieson T, Marks L, Anscher MS, Prosnitz LR. Radiotherapy alone for medically inoperable stage I non-small-cell lung cancer: the Duke experience. Int J Radiat Oncol Biol Phys 1998;40:149–54. [9] Haasbeek CJ, Senan S, Smit EF, Paul MA, Slotman BJ, Lagerwaard FJ. Critical review of nonsurgical treatment options for stage I non-small cell lung cancer. Oncologist 2008;13:309–19. [10] Fuss M, Salter BJ, Rassiah P, Cheek D, Cavanaugh SX, Herman TS. Repositioning accuracy of a commercially available double-vacuum whole body immobilization system for stereotactic body radiation therapy. Technol Cancer Res Treat 2004;3:59–67. [11] Nevinny-Stickel M, Sweeney RA, Bale RJ, Posch A, Auberger T, Lukas P. Reproducibility of patient positioning for fractionated extracranial stereotactic radiotherapy using a double-vacuum technique. Strahlenther Onkol 2004;180:117–22. [12] Baba F, Shibamoto Y, Tomita N, et al. Stereotactic body radiotherapy for stage I lung cancer, small lung metastasis: evaluation of an immobilization system for suppression of respiratory tumor movement, preliminary results. Radiat Oncol 2009;4:15. [13] Negoro Y, Nagata Y, Aoki T, et al. The effectiveness of an immobilization device in conformal radiotherapy for lung tumor: reduction of respiratory tumor movement and evaluation of the daily setup accuracy. Int J Radiat Oncol Biol Phys 2008;50:889–98. [14] Heinzerling J, Anderson JF, Papiez L, et al. Four-dimensional computed tomography scan analysis of tumor and organ motion at varying levels of abdominal compression during stereotactic treatment of lung and liver. Int J Radiat Oncol Biol Phys 2008;70:1571–8. [15] Underberg RWM, Lagerwaard FJ, Slotman BJ, et al. Use of maximum intensity projections (MIP) for target volume generation in 4DCT scans for lung cancer. Int J Radiat Oncol Biol Phys 2005;63:253–60.
[16] Rietzel E, Liu AK, Doppke KP, et al. Design of 4D treatment planning target volumes. Int J Radiat Oncol Biol Phys 2006;66:287–95. [17] Sixel KE, Ruschin M, Tirona R, Cheung PC. Digital fluoroscopy to quantify lung tumor motion: Potential for patient-specific planning target volumes. Int J Radiat Oncol Biol Phys 2003;57:717–23. [18] Ekberg L, Holmberg O, Wittgren L, Bjelkengren G, Landberg T. What margins should be added to the clinical target volume in radiotherapy treatment planning for lung cancer. Radiother Oncol 1998;48:71–7. [19] Erridge SC, Seppenwoolde Y, Muller S, et al. Portal imaging to assess set-up errors, tumor motion and tumor shrinkage during conformal radiotherapy of non-small cell lung cancer. Radiother Oncol 2003;66:75–85. [20] Plathow C, Ley S, Fink C, et al. Analysis of intrathoracic tumor mobility during whole breathing cycle by dynamic MRI. Int J Radiat Oncol Biol Phys 2004;59:952–9. [21] Weiss E, Wijesooriya K, Vaughn Dill S, Keall PJ. Tumor and normal tissue motion in the thorax during respiration: analysis of volumetric and positional variations using 4DCT. Int J Radiat Oncol Biol Phys 2007;67:297–307. [22] Guckenberger M, Wilbert J, Meyer J, Baier K, Richter A, Flentje M. Is a single respiratory correlated 4D-CT study sufficient for evaluation of breathing motion? Int J Radiat Oncol Biol Phys 2007;67:1352–9. [23] Liu H, Balter P, Tutt T, et al. Assessing respiration-induced tumor motion and internal target volume using four-dimensional computed tomography for radiotherapy of lung cancer. Int J Radiat Oncol Biol Phys 2007;68:531–40. [24] Grills IS, Hugo G, Kestin LL, et al. Image-guided radiotherapy via daily online cone-beam CT substantially reduced margin requirements for stereotactic lung radiotherapy. Int J Radiat Oncol Biol Phys 2008;70:1045–56. [25] Guckenberger M, Meyer J, Wilbert J, et al. Intra-fractional uncertainties in cone-beam CT based image-guided radiotherapy (IGRT) of pulmonary tumors. Radiother Oncol 2007;83:57–64. [26] Purdie TG, Bissonnette JP, Franks K, et al. Cone-beam computed tomography for on-line image guidance of lung stereotactic radiotherapy: localization, verification, and intrafraction tumor position. Int J Radiat Oncol Biol Phys 2007;68: 243–52.