- Email: [email protected]
Partial kilovoltage cone beam computed tomography, complete kilovoltage cone beam computed tomography, and electronic portal images for breast radiation therapy: A dose-comparison study
Practical Radiation Oncology (2015) 5, e521-e529
www.practicalradonc.org
Original Report
Partial kilovoltage cone beam computed tomography, complete kilovoltage cone beam computed tomography, and electronic portal images for breast radiation therapy: A dose-comparison study Houda Bahig MD, Étienne Roussin MSc, Michael Yassa MD, Peter Vavassis MD, Céline Lemaire MD, Laurie Archambault MSc, David H.A. Nguyen MD ⁎ Radiation Oncology Department, Maisonneuve-Rosemont Hospital, Montréal, Quebec, Canada Received 30 August 2014; revised 28 January 2015; accepted 15 February 2015
Abstract Purpose: The purpose of this study was to compare absorbed dose with the treated breast and organs at risks (OARs) with weekly image guidance using electronic portal imaging (EPI), complete kilovoltage cone beam computed tomography (kV CBCT), and partial kV CBCT. Methods and materials: Using a thorax female phantom, we determined absorbed doses to treated and contralateral breast, ipsilateral and contralateral lung, heart, and skin for tangential EPI, complete kV CBCT, and partial kV CBCT. Doses were measured by use of ionization chambers and compared with treatment planning system calculations. With simulation of breast tangential irradiation to a standard dose of 50 Gy in 25 fractions, dose to each organ was measured for each image guidance technique. Results: Use of weekly EPI was associated with a significantly increased dose to the treated breast compared with weekly complete and partial kV CBCT (4.44 ± 0.04 vs 1.00 ± 0.07 vs 0.576 ± 0.003 cGy, respectively). Dose to the contralateral breast, ipsilateral and contralateral lung, heart, and contralateral skin was lower with EPI than with either complete or partial kV CBCT (0.042 ± 0.004 vs 0.36 ± 0.01 vs 0.23 ± 0.01 cGy, 0.06 ± 0.04 vs 0.42 ± 0.02 vs 0.31 ± 0.01 cGy, 0.004 ± 0.002 vs 0.29 ± 0.01 vs 0.22 ± 0.01 cGy, 0.03 ± 0.08 vs 0.36 ± 0.02 vs 0.25 ± 0.01 cGy, and 0.20 ± 0.02 vs 0.80 ± 0.06 vs 0.40 ± 0.03 cGy, respectively). Compared with complete CBCT, the use of partial CBCT allowed dose reductions of 42%, 37%, 27%, 24%, and 28% to the ipsilateral breast, contralateral breast, ipsilateral lung, contralateral lung, and heart, respectively. Additional dose from weekly CBCT was significantly lower than treatment-related scatter dose for all OARs. Conclusions: Use of CBCT was associated with decreased dose to ipsilateral breast and increased dose to all OARs compared with EPI. Significant dose reduction can be achieved with the use of partial CBCT, while generally maintaining image quality. © 2015 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.
Conflicts of interest: None. ⁎ Corresponding author. Maisonneuve-Rosemont Hospital, Radiation Oncology Department, 5305 Boulevard de l’Assomption, Montreal, Quebec, Canada H1T 2M4. E-mail address: [email protected] (D.H.A. Nguyen). http://dx.doi.org/10.1016/j.prro.2015.02.009 1879-8500/© 2015 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.
e522
H. Bahig et al
Introduction Whole breast radiation therapy with or without a boost to the surgical cavity in combination with breastconserving surgery remains the standard treatment for early-stage breast cancer. 1 Recent developments in image guided radiation therapy (IGRT) have led to improvement in the accuracy of radiation delivery in breast cancer by minimizing interfraction and intrafraction variations in patient setup and internal anatomic changes. 2-4 The choice of planning target volume (PTV) margin accounts for setup uncertainties to ensure appropriate coverage of the clinical target volume and includes both a random error component that varies daily and a systematic error component. 5 The introduction of daily pretreatment imaging allows for correction of patient and setup positioning errors, as well as changes in breast shape. Currently, the standard for breast radiation therapy setup verification in breast cancer patients remains the use of tangential electronic portal imaging (EPI) compared with digitally reconstructed radiographs. 6 With this method, the positions of the ribs, sternum, and vertebrae are used as reference points for proper positioning. Although an increasing number of radiation oncology centers now use kilovoltage (kV) daily images, a significant limitation of this method is that the current On-Board Imager (Varian Medical Systems) kV imaging system only allows verification of the isocenter and not necessarily the beam portal relative to the breast anatomy, as in EPI. Compared with EPI, planar kV offers improved image quality along with exposure to a much lower additional radiation dose, as previously shown in phantom studies. 7 Although both EPI and kV planar imaging techniques may offer sufficient accuracy in tangential breast radiation therapy, both present several limitations when a third field is used for boost irradiation, with field-in-field techniques, or in intensity modulated radiation therapy. The use of tangential EPI and kV planar imaging provides mostly 2-dimensional (2D) information, lacks sensitivity to patient rotational deviations, 8,9 and does not detect setup errors in the direction of the tangential beams. 10 Therefore, when nontangential beams are used to improve coverage or for boost volume irradiation, setup errors in that direction may result in an underdosage of the surgical cavity. Although the use of 2D stereoscopic imaging can provide some limited 3-dimensional (3D) information, this method, which is not available in our center, does not allow localization of soft tissue target. In the era of partial breast irradiation, in which larger doses per fraction are delivered to a more precise volume, small changes in position have the potential to significantly affect radiation precision, with a potential for tumor miss. kV cone beam computed tomography (CBCT) imaging is an IGRT technique that consists of reconstruction of 3D volumes from a series of 2D projections using a kV x-ray tube mounted to the megavoltage treatment beam. 11 This
Practical Radiation Oncology: September-October 2015
method has the potential to further reduce setup error and therefore decrease radiation dose to surrounding organs at risk (OARs) while maintaining appropriate coverage of the treatment volume. 12-14 The use of kV CBCT systems has been increasing in several centers, and a growing number of studies report its superiority over 2D imaging for setup error correction in breast cancer radiation therapy. 8,9,15 The main advantages of this technique include accurate 3D anatomy, soft tissue visualization, fast rigid registrations, and enhanced precision of match. However, with the increasing use of IGRT, the main concern is that of an increased absorbed dose to healthy surrounding organs, with potential secondary cancer risk. Although it is widely accepted that high-dose radiation increases the risk of secondary cancer linearly, 16 quantification of cancer risk from low-dose radiation remains unclear. 17 To benefit from the increased accuracy of CBCT-guided radiation and address the concern of radiation dose to OARs, our center has adopted the use of weekly pretreatment partial CBCT since the summer of 2013. To better quantify radiation doses from different image guidance techniques, this study aimed to compare daily and total dose received by the irradiated breast and OARs using tangential EPI, complete kV CBCT, and an in-house partial CBCT protocol for setup correction in breast external beam radiation therapy.
Methods and materials A custom-designed female thorax phantom was built at our institution. The phantom was made with polymethyl methacrylate plates, as well as cork inserts and paraffin to simulate the lungs and the breasts, respectively. The
Figure 1 Points of measurement for each organ: Treated breast (yellow), contralateral breast (purple), lung (blue), and heart (orange) on axial view of planning computed tomography scan of in-house thoracic phantom.
Practical Radiation Oncology: September-October 2015
CBCT dose in breast cancer radiation therapy
e523
Figure 2 (A) Posterior 270° gantry rotation for partial kilovoltage cone beam computed tomography (kV CBCT). (B) Example of sagittal and axial slices of a partial low-dose kV CBCT of the breast for a patient.
phantom was placed supine on the treatment couch, and a planning computed tomography scan (Fig 1) was acquired with a slice thickness of 3 mm, a tube potential of 120 kilovoltage peak (kVp), and a tube current of 140 mA. Absorbed dose was determined for the treated and contralateral breast, ipsilateral and contralateral lung, and heart. Dose measurements were performed for each imaging protocol: tangential EPI, complete gantry rotation low-dose kV CBCT protocol with Elekta X-ray volume Table 1
imaging software package version 4.5, and three-quarter gantry rotation low-dose kV CBCT protocol. With simulation of breast tangential irradiation to a dose of 50 Gy in 25 fractions, the cumulated dose from weekly imaging was measured for every image guidance technique, for each organ. Doses were measured with Farmertype ion chambers compared with Pinnacle treatment planning system calculations (Philips Inc). The different points of measurement for each organ are presented in
Parameters for the imaging techniques used
Tube voltage Tube current per projection Exposure time per projection (ms) Gantry rotation range No. of projections Exposure (mAs) Bow-tie filter FOV (cm) Blade X1 (cm) Blade X2 (cm) Blade Z1 (cm) Blade Z2 (cm)
EPID
Complete CBCT
Partial CBCT
6 MV 2 MU 9 9 5 5
120 kVp 20 mA 20 360° ~ 660 400 F1 41 M20: 27.67 cm × 27.67 cm - isocenter
120 kVp 20 mA 20 258° ~ 475 400 F1 41 M20: 27.67 cm × 27.67 cm - isocenter
CBCT, cone beam computed tomography; EPID, electronic portal imaging device; FOV, field of view; kVp, kilovoltage peak; mAs, milliamperage second.
e524
H. Bahig et al
Practical Radiation Oncology: September-October 2015
Table 2 Total cumulated dose per organ for each image guidance protocol using a standard fractionation of 50 Gy in 25 fractions for tangential irradiation weekly IGRT Total cumulated dose from weekly IGRT Treatment (cGy) Treated breast Total One acquisition Contralateral breast Total One acquisition Ipsilateral lung Total One acquisition Contralateral lung Total One acquisition Heart Total One acquisition Ipsilateral skin Internal, total One acquisition External, total One acquisition Contralateral skin Internal, total One acquisition External, total One acquisition
EPID (cGy)
Complete CBCT (cGy)
Partial CBCT (cGy)
5000 ± 60 -
22.2 ± 0.2 4.44 ± 0.04
5.0 ± 0.4 1.00 ± 0.07
2.88 ± 0.02 0.576 ± 0.003
24.0 ± 4.0 -
0.2 ± 0.02 0.042 ± 0.004
1.80 ± 0.06 0.36 ± 0.01
1.16 ± 0.04 0.23 ± 0.01
61.0 ± 0.3 -
0.3 ± 0.2 0.06 ± 0.04
2.1 ± 0.1 0.42 ± 0.02
1.56 ± 0.04 0.31 ± 0.01
6.0 ± 0.5 -
0.02 ± 0.01 0.004 ± 0.002
1.48 ± 0.06 0.29 ± 0.01
1.12 ± 0.02 0.22 ± 0.01
26.0 ± 0.1 -
0.15 ± 0.4 0.03 ± 0.08
1.8 ± 0.1 0.36 ± 0.02
1.26 ± 0.02 0.25 ± 0.01
11.5 ± 0.5 2.3 ± 0.1 15.0 ± 1 3.0 ± 0.2
1.5 ± 0.1 0.30 ± 0.02 1.5 ± 0.1 0.30 ± 0.02
1.0 ± 0.1 0.20 ± 0.02 1.5 ± 0.1 0.30 ± 0.02
1.5 ± 0.1 0.30 ± 0.02 0.50 ± 0.05 0.10 ± 0.01
4.0 ± 0.5 0.80 ± 0.06 4.0 ± 0.5 0.80 ± 0.06
2.0 ± 0.2 0.40 ± 0.03 2.0 ± 0.2 0.40 ± 0.03
26,000 ± 131 34,000 ± 171 340 ± 17 110 ± 6 -
CBCT, cone beam computed tomography; EPID, electronic portal imaging device; IGRT, image guided radiation therapy.
Fig 2. For the treated and the contralateral breast, dose was measured at the center of a simulated pair of tangential fields. For each lung, the dose was measured at 5 different anteroposterior points of measurement from the chest wall and at a lateral depth of 7 cm (center of the lung) (Fig 1). The distance from the surface to the chest wall in the axis of measurement (corresponding to breast tissue) was
4.7 cm. Similarly, dose to the heart was measured at 4 points that corresponded to the expected anatomic location of the heart. To evaluate skin surface, Gafchromic EBT3 films (Ashland Inc) were calibrated with a 120-kVp orthovoltage beam on the internal and external surfaces of the ipsilateral and contralateral breast of the phantom (Fig 1). Three pieces of film (4 × 4 cm 2) were irradiated at each
Figure 3 Dose per organ for a single acquisition for each imaging protocol: electronic portal imaging (EPI), partial kilovoltage cone beam computed tomography (kV CBCT), and complete kV CBCT. Contra, contralateral; Ipsi, ipsilateral.
Practical Radiation Oncology: September-October 2015
CBCT dose in breast cancer radiation therapy
e525
Figure 4 Total dose at different phantom depths from weekly image guided radiation therapy (IGRT) and treatment-related scatter dose (50 Gy in 25 fractions) for ipsilateral and contralateral lungs. CBCT, cone beam computed tomography; EPI, electronic portal imaging.
calibration dose and were scanned 3 times to average out the different uncertainties associated with low-dose irradiation and film scanning. The calibration curve was generated in FilmQA Pro (Ashland Inc) by use of the triple-channel method. The film response to dose rate and incidence angle was investigated to quantify the uncertainty on the CBCT dose readings. Parameters used for each imaging technique are detailed in Table 1. Both complete and partial low-dose CBCT parameters were determined empirically with a progressive dose decrease to the lowest dose necessary for adequate clinical visualization of the anatomy. Low-dose partial CBCT was in fact adopted in 2 steps: First, the range of projection was reduced, and then other dose parameters were empirically decreased. Partial CBCT was initially investigated in the context of gantry and couch collision problems that had been encountered. In our experience, three-quarter rotation generally allowed for adequate rotation without gantry collision for all patients. Precision of displacements achieved with three-quarter rotation CBCT was compared with complete CBCT in phantom studies. Automatic registration was found to be as precise as with partial CBCT. CBCT dose reduction was achieved empirically based on a multistep process. First, the effect of the bow-tie filter (F1) on the image quality was compared with F0 filter (no filter effect) by use of Catphan 600 phantom (The Phantom Laboratory Inc). No significant decrease in image quality (signal-to-noise ratio,
spatial resolution, and low-contrast detection) occurred as a function of dose decrease. The computed tomography dose index was therefore reduced by a factor of 1.6. Image quality and precision of displacement of various imaging parameters (kilovoltage, milliamperage) were compared with those of baseline parameters provided by the manufacturer. These measurements were based on Catphan 600 phantom studies. In clinical practice, image acquisition parameters were gradually lowered to those used in the phantom study. Feedback from radiation therapists was obtained to verify the quality of the images and of the surface, bony, and surgical cavity match. Final parameters of complete kV CBCT included a field of view of 41 cm, a field size of 27.7 × 27.7 cm, 120-kVp tube voltage, an average of 660 projections, and use of a bow-tie filter. Partial kV CBCT differed only in the use of a 258° gantry rotation and approximately 450 projections (Fig 2). The number of projections was the average number of projections for the measure sets and was determined by the imaging system. A small difference in the number of projections using the same acquisition parameters could be observed if the acquisition was made several times in a row as a consequence of the gantry speed variation inherent to the imaging system. To be able to measure the absorbed dose with the ionization chambers, the beam quality must be determined for the CBCT settings used clinically in breast treatments. The half-value layer of the XVI unit was measured and
e526
H. Bahig et al
Practical Radiation Oncology: September-October 2015
Figure 5 Total dose at different phantom depths from weekly image guided radiation therapy (IGRT) and treatment-related scatter dose (50 Gy in 25 fractions) for the heart. CBCT, cone beam computed tomography; EPI, electronic portal imaging.
was found to be 0.389 mm Cu. The effective energy, Eeff, was calculated from the half-value layer with a polynomial fitting of published data. 18 The corresponding Nx was interpolated from a fitting of the National Research Council calibration coefficients of an air-calibrated PTW Farmer TN30013 chamber. In accordance with the Institution of Physics and Engineering in Medicine and Biology Code of Practice, 19 the 5 NE2571 cylindrical chambers used in this study were cross-calibrated in air against the National Research Council calibrated chamber.
Results Cumulated doses from weekly IGRT and the dose from single-image acquisition for each method are shown in Table 2 and Fig 3. Table 2 presents point dose to the Table 3
treated and contralateral breast; mean dose to the ipsilateral lung, contralateral lung, and heart; and mean dose to the internal and external surface of both breasts. Use of EPI was associated with a significantly increased dose to the ipsilateral breast and skin compared with complete and partial kV CBCT (4.44 ± 0.04 vs 1.00 ± 0.07 and 0.576 ± 0.003 cGy; 2.7 ± 0.1 vs 0.30 ± 0.02 vs 0.25 ± 0.02 cGy [external and internal mean] per fraction, respectively). Doses to the contralateral breast, heart, ipsilateral and contralateral lung, and contralateral skin were lower with EPI than with either kV CBCT protocol (0.042 ± 0.004 vs 0.36 ± 0.01 vs 0.23 ± 0.01 cGy, 0.06 ± 0.04 vs 0.42 ± 0.02 vs 0.31 ± 0.01 cGy, 0.004 ± 0.002 vs 0.29 ± 0.01 vs 0.22 ± 0.01 cGy, 0.03 ± 0.08 vs 0.36 ± 0.02 vs 0.25 ± 0.01 cGy, and 0.20 ± 0.02 vs 0.80 ± 0.06 vs 0.40 ± 0.03 cGy per fraction, respectively). Fig 4 shows doses measured at each depth for the ipsilateral lung, contralateral lung, and heart.
Summary of reported radiation doses to organs from thoracic kV CBCT
Dose per fraction Kan et al 35 (2008) Low-dose CBCT Alvarado et al 10 (2013) Low-dose CBCT Low-dose 10-ms CBCT CBCT “under breast” Current study Partial CBCT
Ipsilateral breast (cGy)
Contralateral breast (cGy)
1.05 ± 0.04 0.76 ± 0.05 0.38 ± 0.03 0.07 ± 0.02 0.5 ± 0.01
Ipsilateral lung (cGy)
Contralateral lung (cGy)
1.17 ± 0.28
Heart (cGy)
1.52 ± 0.10
0.75 ± 0.05 0.37 ± 0.03 0.11 ± 0.04
0.77 ± 0.05 0.37 ± 0.03 0.29 ± 0.03
0.78 ± 0.05 0.37 ± 0.03 0.30 ± 0.03
1.04 ± 0.07 0.52 ± 0.04 0.24 ± 0.03
0.23 ± 01
0.31 ± 01
0.22 ± 01
0.25 ± 01
CBCT, cone beam computed tomography; kV CBCT, kilovoltage cone beam computed tomography.
Practical Radiation Oncology: September-October 2015
Compared with complete CBCT, the use of partial CBCT allowed a dose reduction at all depths, with mean decreases of 42%, 37%, 27%, 24%, and 28% to the ipsilateral breast, contralateral breast, ipsilateral lung, contralateral lung, and heart, respectively. Mean surface dose reduction achieved with partial kV CBCT was 17% for ipsilateral skin and 50% for contralateral skin. Treatment-related scatter dose was significantly higher than any weekly IGRT-related dose for all organs and at all depths with the exception of the posterior contralateral lung (starting at 14 cm depth), for which treatment-related dose was equal to partial CBCT dose. (See Fig 5.)
Discussion Use of 2D image guidance for breast external beam radiation therapy is limited to 2D bony anatomy information, does not allow soft tissue target localization, and lacks sensitivity to patient rotational deviations and setup errors in the direction of the tangential beam. 8,9 Several studies have in fact suggested an increased accuracy of radiation therapy delivery with this method. 8,13,15 Several previous studies have evaluated the role of kV CBCT for IGRT in breast cancer. A study by White et al 12 evaluated the role of kV CBCT guidance for 20 patients treated with accelerated partial breast irradiation with a PTV margin of 1 cm and showed that the addition of CBCT would allow a reduction of PTV margin by a factor of 2. Topolnjak et al 8 compared differences in setup errors using CBCT and EPI for 20 breast cancer patients treated with 2 opposing fields with a simultaneous integrated boost technique (28-fraction treatment) and showed that EPI registration underestimated the actual bony anatomy setup error by 20% to 50% and that CBCT introduction decreased this uncertainty significantly. On the other hand, Fatunase et al 13 assessed residual setup error after 2D alignment based on bony anatomy and the dosimetric impact of this residual error for 10 patients undergoing accelerated partial breast irradiation. The residual error in soft tissue was found to typically be less than 5 mm, and the dosimetric consequences were minimal, with a mean difference in the PTV and clinical target volume receiving 95% of the prescribed dose of 1% and 4%, respectively. Investigators recognized, however, the limitation of a small patient cohort and suggested that CBCT may have an added clinical benefit in a subset of patients with large breasts or patients in need of tight dosimetric margins. Kim et al 20 published their experience with the use of CBCT to visualize surgical clips in patients undergoing whole breast irradiation. They reported markedly improved beam targeting and reduced PTV margin from 1 to 0.6 cm when combining CBCT visualization of surgical clips, corrections for breathing, changes in the tumor cavity, and other sources of error. Although studies have shown
CBCT dose in breast cancer radiation therapy
e527
increased accuracy with the use of CBCT compared with tangential 2D imaging in breast setup corrections, no previous papers have compared CBCT accuracy with the combined use of 2D stereoscopic imaging and a beam’s-eye view image that could provide both rotational and translational information. Comparison of CBCT versus stereoscopic 2D imaging in terms of accuracy of registration and the delivered dose to OARs should be evaluated further in future studies. In the context of increasing use of IGRT, there are growing concerns about risks of secondary malignancies. As an example, several epidemiologic studies have estimated an increased risk of breast cancer in patients undergoing mammograms as part of screening programs (doses on the order of 1-5 mGy per mammogram). 21-23 However, although it is accepted that high-dose radiation linearly increases the risk of secondary cancer, 16 quantification of cancer risk from low-dose radiation remains unclear. 17 On the basis of extrapolation from high-dose radiation data, it is currently assumed that the linear nonthreshold (LNT) model applies to low-dose radiation. 24 LNT implies a proportional relation between dose received and risk of cancer. Any dose is therefore assumed to carry a risk. Several studies have applied this model to estimate lifetime attributable risk of cancer from low-dose diagnostic imaging 25-28; however, there is controversy as to whether this model is valid, and several alternative models for low-dose radiation effect have been reported. In fact, although the threshold model supports that there is a harmless level of radiation exposure not associated with an increased cancer risk, 25-28 the adaptive response model supports that low-dose radiation may be protective, 29-31 and a recent animal study even suggested a secondary enhanced resistance to larger radiation doses with IGRT. 32 On the other hand, the hypersensitivity model supports that a much greater cancer risk than that estimated by the LNT model is associated with low-dose radiation through various mechanisms. 33,34 The relationship between low dose and cancer risk therefore remains controversial, and lacking a robust low-dose-effect model, the LNT model is used according to the precautionary principle. The advantage of CBCT is the increased precision of delivery of high-dose radiation at the cost of diffuse low-dose radiation to all organs of the thorax. Studies have reported the advantages, in terms of reduction of setup errors and reduction of PTV margins, with the use of CBCT image guidance compared with EPI. This has the potential impact to reduce high to intermediate dose to OARs immediately adjacent to the target volume. In fact, a better match and PTV reduction with the use of a more precise IGRT technique may allow a dose reduction to OARs. Efforts to reduce maximal CBCT dose to achieve the best compromise of an adequate match without an unnecessary supplementary dose should be undertaken. In this study, we propose a partial CBCT protocol that allows for significant reduction of dose to the OAR while offering
e528
H. Bahig et al
the advantages of a 3D kV image match. Partial CBCT offers the same advantages of complete CBCT, allowing for surface, clip, and bony or soft tissue registration, as well as correction for rotational errors. Adequate visualization of the seroma was a criterion used by our radiation therapists at the time of clinical evaluation of low-dose partial CBCT. Even though the parameters used in our study allowed adequate visualization of most surgical cavities, one can assume that there may be cases in which the image quality of partial CBCT may not be sufficient for adequate visualization of the seroma (eg, very large patient geometry). Few other studies have examined radiation dose to OARs from thoracic CBCT protocols. 10,35 A recent German study 36 investigated various prone and supine kV CBCT protocols by altering tube current, exposure time, range of projection views, and fields of view. Although they do not report measured doses to OARs, they demonstrated that a low-dose partial CBCT protocol with a reduced field of view, a decreased range of projection views (180°), a tube current of 20 mA, and an exposure time of 32 ms allowed for adequate contour accuracy, elimination of the offset CBCT isocenter procedure, and a 4.3-fold dose reduction. Kan et al 35 evaluated radiation doses from pelvic, head and neck, and thorax kV CBCT using a low-dose-mode CBCT. Thoracic CBCT delivered doses on the order of 1 cGy to the breast, 1.2 cGy to the lung, and 1.5 cGy to the heart. These reported doses are higher than those reported in our study and the study by Alvarado et al. 10 In the latter study, radiation doses to breast, heart, lung, and skin using 4 different protocols were reported: a standard orthogonal kV image protocol, a standard low-dose thorax CBCT protocol, and 2 dose-reduction protocols (an “under breast” protocol that consisted of partial gantry rotation limited to the posterior side of the patient and a “low-dose thorax 10-ms protocol” that consisted of a reduced current time. Reduction of beam rotation range or scan exposure time significantly reduced the dose to all organs investigated. In our study, doses to OARs obtained with our posterior three-quarter gantry rotation kV CBCT compared well with the lowest dose protocol (under breast) in the study by Alvarado et al 10 (Table 3). Interestingly, we found that the additional dose from partial CBCT was very small compared with treatment-related scatter dose for all OARs, with the exception of the posterior contralateral lung. Strategies to reduce kV CBCT dose are multiple and should aim to reduce maximal radiation exposure without impairing the precision of the match. Protocols to reduce CBCT frequency contribute to further lower the dose to OAR but require appropriate patient selection. In our center, the kV CBCT protocol consists of daily CBCT for the first 3 days of treatment followed by a movement analysis; a weekly CBCT protocol is adopted if displacements are below 1 cm and rotations below 3°, and an individualized CBCT protocol is adopted for more important movements. Image acquisition parameters
Practical Radiation Oncology: September-October 2015
such as tube voltage and exposure (both tube current and exposure time) can be optimized to achieve acceptable image quality, thereby avoiding unnecessary dose. Similarly, a standard field of view of 41 cm was used in this phantom study, but individualization of the field of interest for each patient with acquisition limited to the breast and minimal bony anatomy could reduce dose to healthy organs. In this study, a partial posterior 270° gantry rotation (with limitation of projection number) was associated with up to a 43% dose reduction to OARs. An important consideration in the comparison of megavoltage and kV images is that megavoltage EPIs will give maximum dose at depth, whereas kV images give the maximum dose on the tissue surface. As a consequence, the surface dose reported for kV CBCT will be the dose received over a depth of 1 cm (therefore reaching the breast), because there is a dose plateau between 0 and 1 cm. 37 In addition, with kV energies, there is a differential absorption of radiation in bone of up to 4 times the dose in soft tissue. Our phantom does not have a rib cage, and thus, the dose measured in the lungs in this study is higher than the dose that would have been expected at the same depth should there have been a rib cage. In a real patient, kV dose to the lungs could be further reduced because of extra absorption in the rib bones. In conclusion, use of CBCT was associated with a decreased dose to the ipsilateral breast and an increased dose to all OARs compared with EPI. The additional kV CBCT dose was associated with better treatment precision and therefore potential reduced dose to OARs from radiation treatment. Compared with complete CBCT, partial CBCT was associated with a significant dose reduction of up to 50% to OARs and irradiated breast, while generally maintaining image quality. Partial kV CBCT dose was small compared with treatment-related scatter dose.
References 1. Benda RK, Mendenhall NP, Lind DS, et al. Breast-conserving therapy (BCT) for early-stage breast cancer. J Surg Oncol. 2004;85:14-27. 2. Djajaputra D, Li S. Real-time 3D surface-image-guided beam setup in radiotherapy of breast cancer. Med Phys. 2005;32:65-75. 3. Fuss M, Salter BJ, Cavanaugh SX, et al. Daily ultrasound-based image-guided targeting for radiotherapy of upper abdominal malignancies. Int J Radiat Oncol Biol Phys. 2004;59:1245-1256. 4. Quinn A, Holloway L, Cutajar D, Hardcastle N, Rosenfeld A, Metcalfe P. Megavoltage cone beam CT near surface dose measurements: Potential implications for breast radiotherapy. Med Phys. 2011;38:6222-6227. 5. McKenzie AL, van Herk M, Mijnheer B. The width of margins in radiotherapy treatment plans. Phys Med Biol. 2000;45:3331-3342. 6. van der Laan HP, Hurkmans CW, Kuten A, Westenberg HA, on behalf of the EORTC-ROG Breast Working Party. Current technological clinical practice in breast radiotherapy: Results of a survey in EORTC-Radiation Oncology Group affiliated institutions. Radiother Oncol. 2010;94:280-285.
Practical Radiation Oncology: September-October 2015 7. Walter C, Boda-Heggemann J, Wertz H, et al. Phantom and in-vivo measurements of dose exposure by image-guided radiotherapy (IGRT): MV portal images vs. kV portal images vs. cone-beam CT. Radiother Oncol. 2007;85:418-423. 8. Topolnjak R, Sonke JJ, Nijkamp J, et al. Breast patient setup error assessment: Comparison of electronic portal image devices and cone-beam computed tomography matching results. Int J Radiat Oncol Biol Phys. 2010;78:1235-1243. 9. Gierga DP, Riboldi M, Turcotte JC, et al. Comparison of target registration errors for multiple image-guided techniques in accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys. 2008;70:1239-1246. 10. Alvarado R, Booth JT, Bromley RM, Gustafsson HB. An investigation of image guidance dose for breast radiotherapy. J Appl Clin Med Phys. 2013;14:4085. 11. Groh BA, Siewerdsen JH, Drake DG, Wong JW, Jaffray DA. A performance comparison of flat-panel imager-based MV and kV cone-beam CT. Med Phys. 2002;29:967-975. 12. White EA, Cho J, Vallis KA, et al. Cone beam computed tomography guidance for setup of patients receiving accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys. 2007;68:547-554. 13. Fatunase T, Wang Z, Yoo S, et al. Assessment of the residual error in soft tissue setup in patients undergoing partial breast irradiation: Results of a prospective study using cone-beam computed tomography. Int J Radiat Oncol Biol Phys. 2008;70:1025-1034. 14. De Puysseleyr A, Veldeman L, Bogaert E, De Wagter C, De Neve W. Optimizing image acquisition settings for cone-beam computed tomography in supine and prone breast radiotherapy. Radiother Oncol. 2011;100:227-230. 15. Jaffray DA, Siewerdsen JH, Wong JW, Martinez AA. Flat-panel cone-beam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys. 2002;53:1337-1349. 16. Preston DL, Pierce DA, Shimizu Y, et al. Effect of recent changes in atomic bomb survivor dosimetry on cancer mortality risk estimates. Radiat Res. 2004;162:377-389. 17. Tubiana M, Feinendegen LE, Yang C, Kaminski JM. The linear no-threshold relationship is inconsistent with radiation biologic and experimental data. Radiology. 2009;251:13-22. 18. Khan FM. The Physics of Radiation Therapy. Philadelphia, PA: Lippincott Williams & Wilkins. 2010. 19. Aukett RJ, Burns JE, Greener AG, et al. Addendum to the IPEMB code of practice for the determination of absorbed dose for x-rays below 300 kV generating potential (0.035 mm Al-4 mm Cu HVL). Phys Med Biol. 2005;50:2739-2748. 20. Kim LH, Wong J, Yan D. On-line localization of the lumpectomy cavity using surgical clips. Int J Radiat Oncol Biol Phys. 2007;69: 1305-1309. 21. Berrington de Gonzalez A, Berg CD, Visvanathan K, Robson M. Estimated risk of radiation-induced breast cancer from mammographic screening for young BRCA mutation carriers. J Natl Cancer Inst. 2009;101:205-209.
CBCT dose in breast cancer radiation therapy
e529
22. Yaffe MJ, Mainprize JG. Risk of radiation-induced breast cancer from mammographic screening [published correction appears in Radiology. 2012;264:306]. Radiology. 2011;258:98-105. 23. de Gelder R, Draisma G, Heijnsdijk EA, de Koning HJ. Population-based mammography screening below age 50: Balancing radiation-induced vs prevented breast cancer deaths. Br J Cancer. 2011;104:1214-1220. 24. Cohen BL. The cancer risk from low level radiation. In: Tack D, Kalra MK, Gevenois PA, eds. Radiation Dose From Multidetector CT. Berlin, Germany: Springer-Verlag; 2012. p. 61-79. 25. Einstein AJ, Henzlova MJ, Rajagopalan S. Estimating risk of cancer associated with radiation exposure from 64-slice computed tomography coronary angiography. JAMA. 2007;298:317-323. 26. Muirhead CR, O'Hagan JA, Haylock RG, et al. Mortality and cancer incidence following occupational radiation exposure: Third analysis of the National Registry for Radiation Workers. Br J Cancer. 2009;100:206-212. 27. Faletra FF, D'Angeli I, Klersy C, et al. Estimates of lifetime attributable risk of cancer after a single radiation exposure from 64-slice computed tomographic coronary angiography. Heart. 2010;96:927-932. 28. Smith-Bindman R, Lipson J, Marcus R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med. 2009;169:2078-2086. 29. Tapio S, Jacob V. Radioadaptive response revisited. Radiat Environ Biophys. 2007;46:1-12. 30. Feinendegen LE. Evidence for beneficial low level radiation effects and radiation hormesis. Br J Radiol. 2005;78:3-7. 31. Ina Y, Tanooka H, Yamada T, Sakai K. Suppression of thymic lymphoma induction by life-long low-dose-rate irradiation accompanied by immune activation in C57BL/6 mice. Radiat Res. 2005;163:153-158. 32. Grdina DJ, Murley JS, Miller RC, et al. A survivin-associated adaptive response in radiation therapy. Cancer Res. 2013;73:4418-4428. 33. Mullenders L, Atkinson M, Paretzke H, Sabatier L, Bouffler S. Assessing cancer risks of low-dose radiation. Nat Rev Cancer. 2009;9:596-604. 34. Ojima M, Ban N, Kai M. DNA double-strand breaks induced by very low X-ray doses are largely due to bystander effects. Radiat Res. 2008;170:365-371. 35. Kan MW, Leung LH, Wong W, Lam N. Radiation dose from cone beam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys. 2008;70:272-279. 36. De Puysseleyr A, Mulliez T, Gulyban A, et al. Improved cone-beam computed tomography in supine and prone breast radiotherapy: Surface reconstruction, radiation exposure, and clinical workflow. Strahlenther Onkol. 2013;189:945-950. 37. Downes P, Jarvis R, Radu E, Kawrakow I, Spezi E. Monte Carlo simulation and patient dosimetry for a kilovoltage cone-beam CT unit. Med Phys. 2009;36:4156-4167.
www.practicalradonc.org
Original Report
Partial kilovoltage cone beam computed tomography, complete kilovoltage cone beam computed tomography, and electronic portal images for breast radiation therapy: A dose-comparison study Houda Bahig MD, Étienne Roussin MSc, Michael Yassa MD, Peter Vavassis MD, Céline Lemaire MD, Laurie Archambault MSc, David H.A. Nguyen MD ⁎ Radiation Oncology Department, Maisonneuve-Rosemont Hospital, Montréal, Quebec, Canada Received 30 August 2014; revised 28 January 2015; accepted 15 February 2015
Abstract Purpose: The purpose of this study was to compare absorbed dose with the treated breast and organs at risks (OARs) with weekly image guidance using electronic portal imaging (EPI), complete kilovoltage cone beam computed tomography (kV CBCT), and partial kV CBCT. Methods and materials: Using a thorax female phantom, we determined absorbed doses to treated and contralateral breast, ipsilateral and contralateral lung, heart, and skin for tangential EPI, complete kV CBCT, and partial kV CBCT. Doses were measured by use of ionization chambers and compared with treatment planning system calculations. With simulation of breast tangential irradiation to a standard dose of 50 Gy in 25 fractions, dose to each organ was measured for each image guidance technique. Results: Use of weekly EPI was associated with a significantly increased dose to the treated breast compared with weekly complete and partial kV CBCT (4.44 ± 0.04 vs 1.00 ± 0.07 vs 0.576 ± 0.003 cGy, respectively). Dose to the contralateral breast, ipsilateral and contralateral lung, heart, and contralateral skin was lower with EPI than with either complete or partial kV CBCT (0.042 ± 0.004 vs 0.36 ± 0.01 vs 0.23 ± 0.01 cGy, 0.06 ± 0.04 vs 0.42 ± 0.02 vs 0.31 ± 0.01 cGy, 0.004 ± 0.002 vs 0.29 ± 0.01 vs 0.22 ± 0.01 cGy, 0.03 ± 0.08 vs 0.36 ± 0.02 vs 0.25 ± 0.01 cGy, and 0.20 ± 0.02 vs 0.80 ± 0.06 vs 0.40 ± 0.03 cGy, respectively). Compared with complete CBCT, the use of partial CBCT allowed dose reductions of 42%, 37%, 27%, 24%, and 28% to the ipsilateral breast, contralateral breast, ipsilateral lung, contralateral lung, and heart, respectively. Additional dose from weekly CBCT was significantly lower than treatment-related scatter dose for all OARs. Conclusions: Use of CBCT was associated with decreased dose to ipsilateral breast and increased dose to all OARs compared with EPI. Significant dose reduction can be achieved with the use of partial CBCT, while generally maintaining image quality. © 2015 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.
Conflicts of interest: None. ⁎ Corresponding author. Maisonneuve-Rosemont Hospital, Radiation Oncology Department, 5305 Boulevard de l’Assomption, Montreal, Quebec, Canada H1T 2M4. E-mail address: [email protected] (D.H.A. Nguyen). http://dx.doi.org/10.1016/j.prro.2015.02.009 1879-8500/© 2015 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.
e522
H. Bahig et al
Introduction Whole breast radiation therapy with or without a boost to the surgical cavity in combination with breastconserving surgery remains the standard treatment for early-stage breast cancer. 1 Recent developments in image guided radiation therapy (IGRT) have led to improvement in the accuracy of radiation delivery in breast cancer by minimizing interfraction and intrafraction variations in patient setup and internal anatomic changes. 2-4 The choice of planning target volume (PTV) margin accounts for setup uncertainties to ensure appropriate coverage of the clinical target volume and includes both a random error component that varies daily and a systematic error component. 5 The introduction of daily pretreatment imaging allows for correction of patient and setup positioning errors, as well as changes in breast shape. Currently, the standard for breast radiation therapy setup verification in breast cancer patients remains the use of tangential electronic portal imaging (EPI) compared with digitally reconstructed radiographs. 6 With this method, the positions of the ribs, sternum, and vertebrae are used as reference points for proper positioning. Although an increasing number of radiation oncology centers now use kilovoltage (kV) daily images, a significant limitation of this method is that the current On-Board Imager (Varian Medical Systems) kV imaging system only allows verification of the isocenter and not necessarily the beam portal relative to the breast anatomy, as in EPI. Compared with EPI, planar kV offers improved image quality along with exposure to a much lower additional radiation dose, as previously shown in phantom studies. 7 Although both EPI and kV planar imaging techniques may offer sufficient accuracy in tangential breast radiation therapy, both present several limitations when a third field is used for boost irradiation, with field-in-field techniques, or in intensity modulated radiation therapy. The use of tangential EPI and kV planar imaging provides mostly 2-dimensional (2D) information, lacks sensitivity to patient rotational deviations, 8,9 and does not detect setup errors in the direction of the tangential beams. 10 Therefore, when nontangential beams are used to improve coverage or for boost volume irradiation, setup errors in that direction may result in an underdosage of the surgical cavity. Although the use of 2D stereoscopic imaging can provide some limited 3-dimensional (3D) information, this method, which is not available in our center, does not allow localization of soft tissue target. In the era of partial breast irradiation, in which larger doses per fraction are delivered to a more precise volume, small changes in position have the potential to significantly affect radiation precision, with a potential for tumor miss. kV cone beam computed tomography (CBCT) imaging is an IGRT technique that consists of reconstruction of 3D volumes from a series of 2D projections using a kV x-ray tube mounted to the megavoltage treatment beam. 11 This
Practical Radiation Oncology: September-October 2015
method has the potential to further reduce setup error and therefore decrease radiation dose to surrounding organs at risk (OARs) while maintaining appropriate coverage of the treatment volume. 12-14 The use of kV CBCT systems has been increasing in several centers, and a growing number of studies report its superiority over 2D imaging for setup error correction in breast cancer radiation therapy. 8,9,15 The main advantages of this technique include accurate 3D anatomy, soft tissue visualization, fast rigid registrations, and enhanced precision of match. However, with the increasing use of IGRT, the main concern is that of an increased absorbed dose to healthy surrounding organs, with potential secondary cancer risk. Although it is widely accepted that high-dose radiation increases the risk of secondary cancer linearly, 16 quantification of cancer risk from low-dose radiation remains unclear. 17 To benefit from the increased accuracy of CBCT-guided radiation and address the concern of radiation dose to OARs, our center has adopted the use of weekly pretreatment partial CBCT since the summer of 2013. To better quantify radiation doses from different image guidance techniques, this study aimed to compare daily and total dose received by the irradiated breast and OARs using tangential EPI, complete kV CBCT, and an in-house partial CBCT protocol for setup correction in breast external beam radiation therapy.
Methods and materials A custom-designed female thorax phantom was built at our institution. The phantom was made with polymethyl methacrylate plates, as well as cork inserts and paraffin to simulate the lungs and the breasts, respectively. The
Figure 1 Points of measurement for each organ: Treated breast (yellow), contralateral breast (purple), lung (blue), and heart (orange) on axial view of planning computed tomography scan of in-house thoracic phantom.
Practical Radiation Oncology: September-October 2015
CBCT dose in breast cancer radiation therapy
e523
Figure 2 (A) Posterior 270° gantry rotation for partial kilovoltage cone beam computed tomography (kV CBCT). (B) Example of sagittal and axial slices of a partial low-dose kV CBCT of the breast for a patient.
phantom was placed supine on the treatment couch, and a planning computed tomography scan (Fig 1) was acquired with a slice thickness of 3 mm, a tube potential of 120 kilovoltage peak (kVp), and a tube current of 140 mA. Absorbed dose was determined for the treated and contralateral breast, ipsilateral and contralateral lung, and heart. Dose measurements were performed for each imaging protocol: tangential EPI, complete gantry rotation low-dose kV CBCT protocol with Elekta X-ray volume Table 1
imaging software package version 4.5, and three-quarter gantry rotation low-dose kV CBCT protocol. With simulation of breast tangential irradiation to a dose of 50 Gy in 25 fractions, the cumulated dose from weekly imaging was measured for every image guidance technique, for each organ. Doses were measured with Farmertype ion chambers compared with Pinnacle treatment planning system calculations (Philips Inc). The different points of measurement for each organ are presented in
Parameters for the imaging techniques used
Tube voltage Tube current per projection Exposure time per projection (ms) Gantry rotation range No. of projections Exposure (mAs) Bow-tie filter FOV (cm) Blade X1 (cm) Blade X2 (cm) Blade Z1 (cm) Blade Z2 (cm)
EPID
Complete CBCT
Partial CBCT
6 MV 2 MU 9 9 5 5
120 kVp 20 mA 20 360° ~ 660 400 F1 41 M20: 27.67 cm × 27.67 cm - isocenter
120 kVp 20 mA 20 258° ~ 475 400 F1 41 M20: 27.67 cm × 27.67 cm - isocenter
CBCT, cone beam computed tomography; EPID, electronic portal imaging device; FOV, field of view; kVp, kilovoltage peak; mAs, milliamperage second.
e524
H. Bahig et al
Practical Radiation Oncology: September-October 2015
Table 2 Total cumulated dose per organ for each image guidance protocol using a standard fractionation of 50 Gy in 25 fractions for tangential irradiation weekly IGRT Total cumulated dose from weekly IGRT Treatment (cGy) Treated breast Total One acquisition Contralateral breast Total One acquisition Ipsilateral lung Total One acquisition Contralateral lung Total One acquisition Heart Total One acquisition Ipsilateral skin Internal, total One acquisition External, total One acquisition Contralateral skin Internal, total One acquisition External, total One acquisition
EPID (cGy)
Complete CBCT (cGy)
Partial CBCT (cGy)
5000 ± 60 -
22.2 ± 0.2 4.44 ± 0.04
5.0 ± 0.4 1.00 ± 0.07
2.88 ± 0.02 0.576 ± 0.003
24.0 ± 4.0 -
0.2 ± 0.02 0.042 ± 0.004
1.80 ± 0.06 0.36 ± 0.01
1.16 ± 0.04 0.23 ± 0.01
61.0 ± 0.3 -
0.3 ± 0.2 0.06 ± 0.04
2.1 ± 0.1 0.42 ± 0.02
1.56 ± 0.04 0.31 ± 0.01
6.0 ± 0.5 -
0.02 ± 0.01 0.004 ± 0.002
1.48 ± 0.06 0.29 ± 0.01
1.12 ± 0.02 0.22 ± 0.01
26.0 ± 0.1 -
0.15 ± 0.4 0.03 ± 0.08
1.8 ± 0.1 0.36 ± 0.02
1.26 ± 0.02 0.25 ± 0.01
11.5 ± 0.5 2.3 ± 0.1 15.0 ± 1 3.0 ± 0.2
1.5 ± 0.1 0.30 ± 0.02 1.5 ± 0.1 0.30 ± 0.02
1.0 ± 0.1 0.20 ± 0.02 1.5 ± 0.1 0.30 ± 0.02
1.5 ± 0.1 0.30 ± 0.02 0.50 ± 0.05 0.10 ± 0.01
4.0 ± 0.5 0.80 ± 0.06 4.0 ± 0.5 0.80 ± 0.06
2.0 ± 0.2 0.40 ± 0.03 2.0 ± 0.2 0.40 ± 0.03
26,000 ± 131 34,000 ± 171 340 ± 17 110 ± 6 -
CBCT, cone beam computed tomography; EPID, electronic portal imaging device; IGRT, image guided radiation therapy.
Fig 2. For the treated and the contralateral breast, dose was measured at the center of a simulated pair of tangential fields. For each lung, the dose was measured at 5 different anteroposterior points of measurement from the chest wall and at a lateral depth of 7 cm (center of the lung) (Fig 1). The distance from the surface to the chest wall in the axis of measurement (corresponding to breast tissue) was
4.7 cm. Similarly, dose to the heart was measured at 4 points that corresponded to the expected anatomic location of the heart. To evaluate skin surface, Gafchromic EBT3 films (Ashland Inc) were calibrated with a 120-kVp orthovoltage beam on the internal and external surfaces of the ipsilateral and contralateral breast of the phantom (Fig 1). Three pieces of film (4 × 4 cm 2) were irradiated at each
Figure 3 Dose per organ for a single acquisition for each imaging protocol: electronic portal imaging (EPI), partial kilovoltage cone beam computed tomography (kV CBCT), and complete kV CBCT. Contra, contralateral; Ipsi, ipsilateral.
Practical Radiation Oncology: September-October 2015
CBCT dose in breast cancer radiation therapy
e525
Figure 4 Total dose at different phantom depths from weekly image guided radiation therapy (IGRT) and treatment-related scatter dose (50 Gy in 25 fractions) for ipsilateral and contralateral lungs. CBCT, cone beam computed tomography; EPI, electronic portal imaging.
calibration dose and were scanned 3 times to average out the different uncertainties associated with low-dose irradiation and film scanning. The calibration curve was generated in FilmQA Pro (Ashland Inc) by use of the triple-channel method. The film response to dose rate and incidence angle was investigated to quantify the uncertainty on the CBCT dose readings. Parameters used for each imaging technique are detailed in Table 1. Both complete and partial low-dose CBCT parameters were determined empirically with a progressive dose decrease to the lowest dose necessary for adequate clinical visualization of the anatomy. Low-dose partial CBCT was in fact adopted in 2 steps: First, the range of projection was reduced, and then other dose parameters were empirically decreased. Partial CBCT was initially investigated in the context of gantry and couch collision problems that had been encountered. In our experience, three-quarter rotation generally allowed for adequate rotation without gantry collision for all patients. Precision of displacements achieved with three-quarter rotation CBCT was compared with complete CBCT in phantom studies. Automatic registration was found to be as precise as with partial CBCT. CBCT dose reduction was achieved empirically based on a multistep process. First, the effect of the bow-tie filter (F1) on the image quality was compared with F0 filter (no filter effect) by use of Catphan 600 phantom (The Phantom Laboratory Inc). No significant decrease in image quality (signal-to-noise ratio,
spatial resolution, and low-contrast detection) occurred as a function of dose decrease. The computed tomography dose index was therefore reduced by a factor of 1.6. Image quality and precision of displacement of various imaging parameters (kilovoltage, milliamperage) were compared with those of baseline parameters provided by the manufacturer. These measurements were based on Catphan 600 phantom studies. In clinical practice, image acquisition parameters were gradually lowered to those used in the phantom study. Feedback from radiation therapists was obtained to verify the quality of the images and of the surface, bony, and surgical cavity match. Final parameters of complete kV CBCT included a field of view of 41 cm, a field size of 27.7 × 27.7 cm, 120-kVp tube voltage, an average of 660 projections, and use of a bow-tie filter. Partial kV CBCT differed only in the use of a 258° gantry rotation and approximately 450 projections (Fig 2). The number of projections was the average number of projections for the measure sets and was determined by the imaging system. A small difference in the number of projections using the same acquisition parameters could be observed if the acquisition was made several times in a row as a consequence of the gantry speed variation inherent to the imaging system. To be able to measure the absorbed dose with the ionization chambers, the beam quality must be determined for the CBCT settings used clinically in breast treatments. The half-value layer of the XVI unit was measured and
e526
H. Bahig et al
Practical Radiation Oncology: September-October 2015
Figure 5 Total dose at different phantom depths from weekly image guided radiation therapy (IGRT) and treatment-related scatter dose (50 Gy in 25 fractions) for the heart. CBCT, cone beam computed tomography; EPI, electronic portal imaging.
was found to be 0.389 mm Cu. The effective energy, Eeff, was calculated from the half-value layer with a polynomial fitting of published data. 18 The corresponding Nx was interpolated from a fitting of the National Research Council calibration coefficients of an air-calibrated PTW Farmer TN30013 chamber. In accordance with the Institution of Physics and Engineering in Medicine and Biology Code of Practice, 19 the 5 NE2571 cylindrical chambers used in this study were cross-calibrated in air against the National Research Council calibrated chamber.
Results Cumulated doses from weekly IGRT and the dose from single-image acquisition for each method are shown in Table 2 and Fig 3. Table 2 presents point dose to the Table 3
treated and contralateral breast; mean dose to the ipsilateral lung, contralateral lung, and heart; and mean dose to the internal and external surface of both breasts. Use of EPI was associated with a significantly increased dose to the ipsilateral breast and skin compared with complete and partial kV CBCT (4.44 ± 0.04 vs 1.00 ± 0.07 and 0.576 ± 0.003 cGy; 2.7 ± 0.1 vs 0.30 ± 0.02 vs 0.25 ± 0.02 cGy [external and internal mean] per fraction, respectively). Doses to the contralateral breast, heart, ipsilateral and contralateral lung, and contralateral skin were lower with EPI than with either kV CBCT protocol (0.042 ± 0.004 vs 0.36 ± 0.01 vs 0.23 ± 0.01 cGy, 0.06 ± 0.04 vs 0.42 ± 0.02 vs 0.31 ± 0.01 cGy, 0.004 ± 0.002 vs 0.29 ± 0.01 vs 0.22 ± 0.01 cGy, 0.03 ± 0.08 vs 0.36 ± 0.02 vs 0.25 ± 0.01 cGy, and 0.20 ± 0.02 vs 0.80 ± 0.06 vs 0.40 ± 0.03 cGy per fraction, respectively). Fig 4 shows doses measured at each depth for the ipsilateral lung, contralateral lung, and heart.
Summary of reported radiation doses to organs from thoracic kV CBCT
Dose per fraction Kan et al 35 (2008) Low-dose CBCT Alvarado et al 10 (2013) Low-dose CBCT Low-dose 10-ms CBCT CBCT “under breast” Current study Partial CBCT
Ipsilateral breast (cGy)
Contralateral breast (cGy)
1.05 ± 0.04 0.76 ± 0.05 0.38 ± 0.03 0.07 ± 0.02 0.5 ± 0.01
Ipsilateral lung (cGy)
Contralateral lung (cGy)
1.17 ± 0.28
Heart (cGy)
1.52 ± 0.10
0.75 ± 0.05 0.37 ± 0.03 0.11 ± 0.04
0.77 ± 0.05 0.37 ± 0.03 0.29 ± 0.03
0.78 ± 0.05 0.37 ± 0.03 0.30 ± 0.03
1.04 ± 0.07 0.52 ± 0.04 0.24 ± 0.03
0.23 ± 01
0.31 ± 01
0.22 ± 01
0.25 ± 01
CBCT, cone beam computed tomography; kV CBCT, kilovoltage cone beam computed tomography.
Practical Radiation Oncology: September-October 2015
Compared with complete CBCT, the use of partial CBCT allowed a dose reduction at all depths, with mean decreases of 42%, 37%, 27%, 24%, and 28% to the ipsilateral breast, contralateral breast, ipsilateral lung, contralateral lung, and heart, respectively. Mean surface dose reduction achieved with partial kV CBCT was 17% for ipsilateral skin and 50% for contralateral skin. Treatment-related scatter dose was significantly higher than any weekly IGRT-related dose for all organs and at all depths with the exception of the posterior contralateral lung (starting at 14 cm depth), for which treatment-related dose was equal to partial CBCT dose. (See Fig 5.)
Discussion Use of 2D image guidance for breast external beam radiation therapy is limited to 2D bony anatomy information, does not allow soft tissue target localization, and lacks sensitivity to patient rotational deviations and setup errors in the direction of the tangential beam. 8,9 Several studies have in fact suggested an increased accuracy of radiation therapy delivery with this method. 8,13,15 Several previous studies have evaluated the role of kV CBCT for IGRT in breast cancer. A study by White et al 12 evaluated the role of kV CBCT guidance for 20 patients treated with accelerated partial breast irradiation with a PTV margin of 1 cm and showed that the addition of CBCT would allow a reduction of PTV margin by a factor of 2. Topolnjak et al 8 compared differences in setup errors using CBCT and EPI for 20 breast cancer patients treated with 2 opposing fields with a simultaneous integrated boost technique (28-fraction treatment) and showed that EPI registration underestimated the actual bony anatomy setup error by 20% to 50% and that CBCT introduction decreased this uncertainty significantly. On the other hand, Fatunase et al 13 assessed residual setup error after 2D alignment based on bony anatomy and the dosimetric impact of this residual error for 10 patients undergoing accelerated partial breast irradiation. The residual error in soft tissue was found to typically be less than 5 mm, and the dosimetric consequences were minimal, with a mean difference in the PTV and clinical target volume receiving 95% of the prescribed dose of 1% and 4%, respectively. Investigators recognized, however, the limitation of a small patient cohort and suggested that CBCT may have an added clinical benefit in a subset of patients with large breasts or patients in need of tight dosimetric margins. Kim et al 20 published their experience with the use of CBCT to visualize surgical clips in patients undergoing whole breast irradiation. They reported markedly improved beam targeting and reduced PTV margin from 1 to 0.6 cm when combining CBCT visualization of surgical clips, corrections for breathing, changes in the tumor cavity, and other sources of error. Although studies have shown
CBCT dose in breast cancer radiation therapy
e527
increased accuracy with the use of CBCT compared with tangential 2D imaging in breast setup corrections, no previous papers have compared CBCT accuracy with the combined use of 2D stereoscopic imaging and a beam’s-eye view image that could provide both rotational and translational information. Comparison of CBCT versus stereoscopic 2D imaging in terms of accuracy of registration and the delivered dose to OARs should be evaluated further in future studies. In the context of increasing use of IGRT, there are growing concerns about risks of secondary malignancies. As an example, several epidemiologic studies have estimated an increased risk of breast cancer in patients undergoing mammograms as part of screening programs (doses on the order of 1-5 mGy per mammogram). 21-23 However, although it is accepted that high-dose radiation linearly increases the risk of secondary cancer, 16 quantification of cancer risk from low-dose radiation remains unclear. 17 On the basis of extrapolation from high-dose radiation data, it is currently assumed that the linear nonthreshold (LNT) model applies to low-dose radiation. 24 LNT implies a proportional relation between dose received and risk of cancer. Any dose is therefore assumed to carry a risk. Several studies have applied this model to estimate lifetime attributable risk of cancer from low-dose diagnostic imaging 25-28; however, there is controversy as to whether this model is valid, and several alternative models for low-dose radiation effect have been reported. In fact, although the threshold model supports that there is a harmless level of radiation exposure not associated with an increased cancer risk, 25-28 the adaptive response model supports that low-dose radiation may be protective, 29-31 and a recent animal study even suggested a secondary enhanced resistance to larger radiation doses with IGRT. 32 On the other hand, the hypersensitivity model supports that a much greater cancer risk than that estimated by the LNT model is associated with low-dose radiation through various mechanisms. 33,34 The relationship between low dose and cancer risk therefore remains controversial, and lacking a robust low-dose-effect model, the LNT model is used according to the precautionary principle. The advantage of CBCT is the increased precision of delivery of high-dose radiation at the cost of diffuse low-dose radiation to all organs of the thorax. Studies have reported the advantages, in terms of reduction of setup errors and reduction of PTV margins, with the use of CBCT image guidance compared with EPI. This has the potential impact to reduce high to intermediate dose to OARs immediately adjacent to the target volume. In fact, a better match and PTV reduction with the use of a more precise IGRT technique may allow a dose reduction to OARs. Efforts to reduce maximal CBCT dose to achieve the best compromise of an adequate match without an unnecessary supplementary dose should be undertaken. In this study, we propose a partial CBCT protocol that allows for significant reduction of dose to the OAR while offering
e528
H. Bahig et al
the advantages of a 3D kV image match. Partial CBCT offers the same advantages of complete CBCT, allowing for surface, clip, and bony or soft tissue registration, as well as correction for rotational errors. Adequate visualization of the seroma was a criterion used by our radiation therapists at the time of clinical evaluation of low-dose partial CBCT. Even though the parameters used in our study allowed adequate visualization of most surgical cavities, one can assume that there may be cases in which the image quality of partial CBCT may not be sufficient for adequate visualization of the seroma (eg, very large patient geometry). Few other studies have examined radiation dose to OARs from thoracic CBCT protocols. 10,35 A recent German study 36 investigated various prone and supine kV CBCT protocols by altering tube current, exposure time, range of projection views, and fields of view. Although they do not report measured doses to OARs, they demonstrated that a low-dose partial CBCT protocol with a reduced field of view, a decreased range of projection views (180°), a tube current of 20 mA, and an exposure time of 32 ms allowed for adequate contour accuracy, elimination of the offset CBCT isocenter procedure, and a 4.3-fold dose reduction. Kan et al 35 evaluated radiation doses from pelvic, head and neck, and thorax kV CBCT using a low-dose-mode CBCT. Thoracic CBCT delivered doses on the order of 1 cGy to the breast, 1.2 cGy to the lung, and 1.5 cGy to the heart. These reported doses are higher than those reported in our study and the study by Alvarado et al. 10 In the latter study, radiation doses to breast, heart, lung, and skin using 4 different protocols were reported: a standard orthogonal kV image protocol, a standard low-dose thorax CBCT protocol, and 2 dose-reduction protocols (an “under breast” protocol that consisted of partial gantry rotation limited to the posterior side of the patient and a “low-dose thorax 10-ms protocol” that consisted of a reduced current time. Reduction of beam rotation range or scan exposure time significantly reduced the dose to all organs investigated. In our study, doses to OARs obtained with our posterior three-quarter gantry rotation kV CBCT compared well with the lowest dose protocol (under breast) in the study by Alvarado et al 10 (Table 3). Interestingly, we found that the additional dose from partial CBCT was very small compared with treatment-related scatter dose for all OARs, with the exception of the posterior contralateral lung. Strategies to reduce kV CBCT dose are multiple and should aim to reduce maximal radiation exposure without impairing the precision of the match. Protocols to reduce CBCT frequency contribute to further lower the dose to OAR but require appropriate patient selection. In our center, the kV CBCT protocol consists of daily CBCT for the first 3 days of treatment followed by a movement analysis; a weekly CBCT protocol is adopted if displacements are below 1 cm and rotations below 3°, and an individualized CBCT protocol is adopted for more important movements. Image acquisition parameters
Practical Radiation Oncology: September-October 2015
such as tube voltage and exposure (both tube current and exposure time) can be optimized to achieve acceptable image quality, thereby avoiding unnecessary dose. Similarly, a standard field of view of 41 cm was used in this phantom study, but individualization of the field of interest for each patient with acquisition limited to the breast and minimal bony anatomy could reduce dose to healthy organs. In this study, a partial posterior 270° gantry rotation (with limitation of projection number) was associated with up to a 43% dose reduction to OARs. An important consideration in the comparison of megavoltage and kV images is that megavoltage EPIs will give maximum dose at depth, whereas kV images give the maximum dose on the tissue surface. As a consequence, the surface dose reported for kV CBCT will be the dose received over a depth of 1 cm (therefore reaching the breast), because there is a dose plateau between 0 and 1 cm. 37 In addition, with kV energies, there is a differential absorption of radiation in bone of up to 4 times the dose in soft tissue. Our phantom does not have a rib cage, and thus, the dose measured in the lungs in this study is higher than the dose that would have been expected at the same depth should there have been a rib cage. In a real patient, kV dose to the lungs could be further reduced because of extra absorption in the rib bones. In conclusion, use of CBCT was associated with a decreased dose to the ipsilateral breast and an increased dose to all OARs compared with EPI. The additional kV CBCT dose was associated with better treatment precision and therefore potential reduced dose to OARs from radiation treatment. Compared with complete CBCT, partial CBCT was associated with a significant dose reduction of up to 50% to OARs and irradiated breast, while generally maintaining image quality. Partial kV CBCT dose was small compared with treatment-related scatter dose.
References 1. Benda RK, Mendenhall NP, Lind DS, et al. Breast-conserving therapy (BCT) for early-stage breast cancer. J Surg Oncol. 2004;85:14-27. 2. Djajaputra D, Li S. Real-time 3D surface-image-guided beam setup in radiotherapy of breast cancer. Med Phys. 2005;32:65-75. 3. Fuss M, Salter BJ, Cavanaugh SX, et al. Daily ultrasound-based image-guided targeting for radiotherapy of upper abdominal malignancies. Int J Radiat Oncol Biol Phys. 2004;59:1245-1256. 4. Quinn A, Holloway L, Cutajar D, Hardcastle N, Rosenfeld A, Metcalfe P. Megavoltage cone beam CT near surface dose measurements: Potential implications for breast radiotherapy. Med Phys. 2011;38:6222-6227. 5. McKenzie AL, van Herk M, Mijnheer B. The width of margins in radiotherapy treatment plans. Phys Med Biol. 2000;45:3331-3342. 6. van der Laan HP, Hurkmans CW, Kuten A, Westenberg HA, on behalf of the EORTC-ROG Breast Working Party. Current technological clinical practice in breast radiotherapy: Results of a survey in EORTC-Radiation Oncology Group affiliated institutions. Radiother Oncol. 2010;94:280-285.
Practical Radiation Oncology: September-October 2015 7. Walter C, Boda-Heggemann J, Wertz H, et al. Phantom and in-vivo measurements of dose exposure by image-guided radiotherapy (IGRT): MV portal images vs. kV portal images vs. cone-beam CT. Radiother Oncol. 2007;85:418-423. 8. Topolnjak R, Sonke JJ, Nijkamp J, et al. Breast patient setup error assessment: Comparison of electronic portal image devices and cone-beam computed tomography matching results. Int J Radiat Oncol Biol Phys. 2010;78:1235-1243. 9. Gierga DP, Riboldi M, Turcotte JC, et al. Comparison of target registration errors for multiple image-guided techniques in accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys. 2008;70:1239-1246. 10. Alvarado R, Booth JT, Bromley RM, Gustafsson HB. An investigation of image guidance dose for breast radiotherapy. J Appl Clin Med Phys. 2013;14:4085. 11. Groh BA, Siewerdsen JH, Drake DG, Wong JW, Jaffray DA. A performance comparison of flat-panel imager-based MV and kV cone-beam CT. Med Phys. 2002;29:967-975. 12. White EA, Cho J, Vallis KA, et al. Cone beam computed tomography guidance for setup of patients receiving accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys. 2007;68:547-554. 13. Fatunase T, Wang Z, Yoo S, et al. Assessment of the residual error in soft tissue setup in patients undergoing partial breast irradiation: Results of a prospective study using cone-beam computed tomography. Int J Radiat Oncol Biol Phys. 2008;70:1025-1034. 14. De Puysseleyr A, Veldeman L, Bogaert E, De Wagter C, De Neve W. Optimizing image acquisition settings for cone-beam computed tomography in supine and prone breast radiotherapy. Radiother Oncol. 2011;100:227-230. 15. Jaffray DA, Siewerdsen JH, Wong JW, Martinez AA. Flat-panel cone-beam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys. 2002;53:1337-1349. 16. Preston DL, Pierce DA, Shimizu Y, et al. Effect of recent changes in atomic bomb survivor dosimetry on cancer mortality risk estimates. Radiat Res. 2004;162:377-389. 17. Tubiana M, Feinendegen LE, Yang C, Kaminski JM. The linear no-threshold relationship is inconsistent with radiation biologic and experimental data. Radiology. 2009;251:13-22. 18. Khan FM. The Physics of Radiation Therapy. Philadelphia, PA: Lippincott Williams & Wilkins. 2010. 19. Aukett RJ, Burns JE, Greener AG, et al. Addendum to the IPEMB code of practice for the determination of absorbed dose for x-rays below 300 kV generating potential (0.035 mm Al-4 mm Cu HVL). Phys Med Biol. 2005;50:2739-2748. 20. Kim LH, Wong J, Yan D. On-line localization of the lumpectomy cavity using surgical clips. Int J Radiat Oncol Biol Phys. 2007;69: 1305-1309. 21. Berrington de Gonzalez A, Berg CD, Visvanathan K, Robson M. Estimated risk of radiation-induced breast cancer from mammographic screening for young BRCA mutation carriers. J Natl Cancer Inst. 2009;101:205-209.
CBCT dose in breast cancer radiation therapy
e529
22. Yaffe MJ, Mainprize JG. Risk of radiation-induced breast cancer from mammographic screening [published correction appears in Radiology. 2012;264:306]. Radiology. 2011;258:98-105. 23. de Gelder R, Draisma G, Heijnsdijk EA, de Koning HJ. Population-based mammography screening below age 50: Balancing radiation-induced vs prevented breast cancer deaths. Br J Cancer. 2011;104:1214-1220. 24. Cohen BL. The cancer risk from low level radiation. In: Tack D, Kalra MK, Gevenois PA, eds. Radiation Dose From Multidetector CT. Berlin, Germany: Springer-Verlag; 2012. p. 61-79. 25. Einstein AJ, Henzlova MJ, Rajagopalan S. Estimating risk of cancer associated with radiation exposure from 64-slice computed tomography coronary angiography. JAMA. 2007;298:317-323. 26. Muirhead CR, O'Hagan JA, Haylock RG, et al. Mortality and cancer incidence following occupational radiation exposure: Third analysis of the National Registry for Radiation Workers. Br J Cancer. 2009;100:206-212. 27. Faletra FF, D'Angeli I, Klersy C, et al. Estimates of lifetime attributable risk of cancer after a single radiation exposure from 64-slice computed tomographic coronary angiography. Heart. 2010;96:927-932. 28. Smith-Bindman R, Lipson J, Marcus R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med. 2009;169:2078-2086. 29. Tapio S, Jacob V. Radioadaptive response revisited. Radiat Environ Biophys. 2007;46:1-12. 30. Feinendegen LE. Evidence for beneficial low level radiation effects and radiation hormesis. Br J Radiol. 2005;78:3-7. 31. Ina Y, Tanooka H, Yamada T, Sakai K. Suppression of thymic lymphoma induction by life-long low-dose-rate irradiation accompanied by immune activation in C57BL/6 mice. Radiat Res. 2005;163:153-158. 32. Grdina DJ, Murley JS, Miller RC, et al. A survivin-associated adaptive response in radiation therapy. Cancer Res. 2013;73:4418-4428. 33. Mullenders L, Atkinson M, Paretzke H, Sabatier L, Bouffler S. Assessing cancer risks of low-dose radiation. Nat Rev Cancer. 2009;9:596-604. 34. Ojima M, Ban N, Kai M. DNA double-strand breaks induced by very low X-ray doses are largely due to bystander effects. Radiat Res. 2008;170:365-371. 35. Kan MW, Leung LH, Wong W, Lam N. Radiation dose from cone beam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys. 2008;70:272-279. 36. De Puysseleyr A, Mulliez T, Gulyban A, et al. Improved cone-beam computed tomography in supine and prone breast radiotherapy: Surface reconstruction, radiation exposure, and clinical workflow. Strahlenther Onkol. 2013;189:945-950. 37. Downes P, Jarvis R, Radu E, Kawrakow I, Spezi E. Monte Carlo simulation and patient dosimetry for a kilovoltage cone-beam CT unit. Med Phys. 2009;36:4156-4167.