- Email: [email protected]
Trajectory modulated prone breast irradiation: A LINAC-based technique combining intensity modulated delivery and motion of the couch
Radiotherapy and Oncology 109 (2013) 475–481
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Trajectory modulated arc therapy
Trajectory modulated prone breast irradiation: A LINAC-based technique combining intensity modulated delivery and motion of the couch Benjamin Fahimian, Victoria Yu, Kathleen Horst, Lei Xing, Dimitre Hristov ⇑ Department of Radiation Oncology, Stanford University, United States
a r t i c l e
i n f o
Article history: Received 14 December 2012 Received in revised form 16 October 2013 Accepted 19 October 2013 Available online 11 November 2013 Keywords: Accelerated partial breast irradiation (APBI) Couch motion IMRT Trajectory modulated arc therapy (TMAT)
a b s t r a c t Purpose: External beam radiation therapy (EBRT) provides a non-invasive treatment alternative for accelerated partial breast irradiation (APBI), however, limitations in achievable dose conformity of current EBRT techniques have been correlated to reported toxicity. To enhance the conformity of EBRT APBI, a technique for conventional LINACs is developed, which through combined motion of the couch, intensity modulated delivery, and a prone breast setup, enables wide-angular coronal arc irradiation of the ipsilateral breast without irradiating through the thorax and contralateral breast. Methods and materials: A couch trajectory optimization technique was developed to determine the trajectories that concurrently avoid collision with the LINAC and maintain the target within the MLC apertures. Inverse treatment planning was performed along the derived trajectory. The technique was experimentally implemented by programming the Varian TrueBeam™ STx in Developer Mode. The dosimetric accuracy of the delivery was evaluated by ion chamber and film measurements in phantom. Results: The resulting optimized trajectory was shown to be necessarily non-isocentric, and contain both translation and rotations of the couch. Film measurements resulted in 93% of the points in the measured two-dimensional dose maps passing the 3%/3 mm Gamma criterion. Preliminary treatment plan comparison to 5-field 3D-conformal, IMRT, and VMAT demonstrated enhancement in conformity, and reduction of the normal tissue V50% and V100% parameters that have been correlated with EBRT toxicity. Conclusions: The feasibility of wide-angular intensity modulated partial breast irradiation using motion of the couch has been demonstrated experimentally on a standard LINAC for the first time. For patients eligible for a prone setup, the technique may enable improvement of dose conformity and associated dose–volume parameters correlated with toxicity. Ó 2013 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 109 (2013) 475–481
Breast conservation therapy consisting of lumpectomy followed by adjuvant radiation therapy is an established treatment for early stage breast cancer. While whole breast irradiation (WBI) is the technique of proven efficacy, alternative techniques using partial breast irradiation of the lumpectomy cavity are under active investigation based on data suggesting that the majority of in breast recurrences occur in the vicinity of the lumpectomy cavity [1–3]. These accelerated partial breast irradiation (APBI) regimens, which additionally allow for a reduction of treatment time to 1 week versus the traditional 6 weeks of WBI, have garnered increased clinical implementation in recent years [4]. Data on the long term efficacy and toxicity of such APBI techniques are still emerging, and ongoing studies such as the Radiation Therapy Oncology Group (RTOG) 0413/National Surgical Adjuvant Breast and Bowel Project (NSABP) B-39 randomized Phase III trial will provide data comparing WBI and APBI [5]. Unlike WBI which aims to uniformly irradiate the breast, partial breast irradiation attempts to conformally irradiate ⇑ Corresponding author. Address: 875 Blake Wilbur Drive, G203, Stanford, CA 94305-5487, United States. E-mail address: [email protected] (D. Hristov). 0167-8140/$ - see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.radonc.2013.10.031
the lumpectomy cavity plus a margin while minimizing dose to the surrounding normal tissue. Since different radiation delivery modalities result in different conformity and normal tissue dose, it is noted that the toxicity of APBI is additionally dependent on the radiation technique [6]. External beam radiotherapy (EBRT) techniques using 3D conformal radiation therapy (3D-CRT) [7] and intensity modulated radiation therapy (IMRT) [8–11] are attractive options for partial breast irradiation because these techniques are non-invasive, widely available and simpler to implement in comparison to brachytherapy alternatives [12]. However, with existing EBRT APBI approaches, unacceptably high rates of moderate-to-severe late normal tissue toxicities have been correlated with several dose– volume parameters [8,11,13]. Specifically, the occurrence of unacceptable cosmesis was correlated with the V100% and V50% [8,11]. Such toxicity can be partially attributed to the fact that current techniques are limited in the achievable dose conformity as they deploy a limited number of tangential field arrangements, which while sparing thoracic normal tissue, inherently irradiate large volumes of breast to moderate and high doses due to limitations in degrees of freedom in the beam arrangements.
476
Trajectory modulated prone breast irradiation
Several alternative methods have been proposed for external beam APBI. These include dedicated stereotactic 60Co breast systems [14], robotic radiosurgery [15], and proton beams [16–18]. While these techniques may allow for variable degrees of improved dosimetry, they require specialized or dedicated systems, which are typically outside the realm of the capabilities of most clinics, and may be questionable in the scope of cost-effective healthcare management. It is thus desirable to develop more conformal types of delivery techniques under conventional LINAC geometry. As such, others have proposed the utilization of a uniform couch rotation under a supine setup [19]. However, no significant dosimetric advantage was achieved relative to a 5-field IMRT technique due to the fact that only a limited uniform continuous couch rotation was modeled (please refer to discussion section for further details). To address the complex dosimetric challenge that partial breast irradiation presents, in this work a delivery method for a conventional LINAC is introduced and developed which combines a nonuniform and non-isocentric couch trajectory, inversely planned intensity modulated delivery, and prone breast setup to enable highly conformal treatments through intensity modulated coronal arc irradiation of the lumpectomy cavity, in a manner avoiding irradiation through the thorax. Specifically, using a method of trajectory based optimization the optimal couch trajectory that concurrently provides wide-angular coverage, avoids collision between the couch and LINAC components, and maintains the target in the beams-eye-view (BEV) of the multi-leaf collimator (MLC), is determined. Inverse planning is used to design an intensity modulated delivery about the trajectory. The feasibility of the proposed trajectory modulated arc therapy (TMAT) delivery technique is experimentally quantified by programming the trajectory modulated delivery on a conventional LINAC. The accuracy of the delivery technique is then quantified using film and ion chamber measurements. While full analysis of multi-patient treatment planning comparisons is beyond the scope of this initial implementation and feasibility work, to demonstrate the type of dosimetry that may be achievable through the technique, a preliminary planning comparison is provided for the delivered case comparing dosimetry of respective implementations using 3D-CRT, IMRT, and en face volumetric modulated arc therapy (VMAT) with avoidance sectors. Method Trajectory optimization The optimization of the delivery trajectory incorporating the motion of the couch entails the determination of the synchronized couch and gantry trajectory which concurrently (1) avoids collisions between all components of the LINAC and the patient, (2) keeps the target in the BEV of the MLC, and (3) provides maximal angular spread of beams. A 4D computer aided design (CAD) model of the LINAC and couch, enabling visualization of the BEV relative to imported patient scans and detection of collisions, was established in SolidWorks™ (Dassault Systèmes SolidWorks Corp., Massachusetts, USA). The trajectory optimization was formulated with the couch modeled as a rigid object, defined as subset of a set of points C in three-dimensional space, C 2 R3 . The transformation of points due to three translational axes and one rotational axis is denoted by combined time dependent transformation Tt ðCÞ : R3 ! R3 . It is noted that the set of points denoted as C includes the patient volume, immobilization device, and target, all encompassed in a safety envelope, as shown in Fig. 1. The LINAC and room geometry is given by a configuration in space denoted by L 2 R3 , which depends on the specific time-dependent orientation of the LINAC.
Subsequently, under a non-collisional couch and LINAC state, the intersection of the two sets is the null set, i.e., C \ L ¼ ;, while a collision is detected if C \ L–;. Besides a non-collisional state between the couch and LINAC components, the trajectory optimization requires that the target be in the field of view of the MLC. A projectional operator is defined as PðMLCÞ : R3 ! R2 and PðPTVÞ : R3 ! R2 , which, at the specific couch angle, projects either the outline of the MLCs or the PTV in a plane orthogonal to the central axis of the beam and passing through the isocenter. This operation yields the point sets of the MLC and PTV projections denoted as P MLC 2 R2 and PPTV 2 R2 , respectively. For a valid beam delivery control point it is required that P MLC \ P PTV –;, that is at least some portion of the PTV should be within the BEV defined by the MLC. The pseudo-code for the trajectory optimization is given in the algorithm in Table 1. All angles are in International Electrotechnical Commission (IEC) 61217 format.
Experimental implementation Currently, no commercial LINAC allows for synchronized motion of the couch during clinical delivery. In order to implement the technique, the Varian TrueBeam™ STx (Serial # 3; Varian Medical Systems, Palo Alto, CA) was utilized in conjunction with Developer Mode capabilities. Under a research environment, Developer Mode allows for custom XML scripting and control of all components of the system, through which, the optimized parameters for motion of the couch, MLC, and beam delivery were programed in this work. The LINAC configuration includes the HD120 MLC, which consists of 60 pairs of leaves, with the central 32 pairs of width of 0.25 cm, flogged symmetrically by 26 pairs of 0.50 cm, and a pair of 0.78 cm MLCs at each edge, resulting in an approximate effective coverage at isocenter of 22 36 cm (18.1 cm maximum extension beyond collimator x-jaw). The maximum supported leaf speed is 2.5 cm/s, and the average leaf transmission averaged across all leaves is <2%. For all implementations in this work, the flattened 6 MV X-ray beam at a dose-rate of 600 MU/min was utilized. The treatment platform included an in-house breast board (enabling elevation of the patient body above the breast opening), coupled with the 4D Exact™ IGRT carbon fiber couch (three degrees of freedom in translation and one in rotation). The Varian Developer Mode schema defines the system control via a Trajectory Model in which all mechanical axis positions and beam delivery parameters are specified with associated trajectory functions. The trajectory functions are defined by the user at discrete cumulative MU control points, in between which a linear
Fig. 1. 4D CAD model of the LINAC, couch, target, and immobilization device utilized for the trajectory optimization.
B. Fahimian et al. / Radiotherapy and Oncology 109 (2013) 475–481
477
Table 1 Trajectory generation algorithm. Prone APBI trajectory generation { SET Gantry ROT = 90°; Couch ROT = 270°; Couch LAT = limit; Couch LONG = limit; Couch VERT such that P MLC \ P PTV –;; SUB-ALGORITHM 1{ IF C \ L–; // No collision between couch or LINAC, // But may not be useful if target outside BEV or MLC limits IF P MLC \ P PTV –; //Target is in beam’s eye view WRITE //Record as delivery control point; END END} Tt ðCÞ : R3 ! R3 ;
// Increment couch either because control point found, // collision detected, or target outside BEV GOTO SUB-ALGORITHM 1 WHILE Couch LONG and ROT < Limit REPEAT for Gantry ROT = 270°}
transformation in the positions and states is interpreted by the system. The experimental implementation of the method proposed here entailed: (1) development of a time-dependent CAD model of the LINAC environment; (2) import of patient CT and contours into the model; (3) optimization of the non-collisional lateral and medial couch trajectories by the algorithm in Table 1, followed by 4D CAD emulation verification; (4) treatment planning optimization by Varian Dose Volume Optimizer, followed by leaf sequencing of the IMRT fluence maps into up to 10 deliverable MLC segments per field; (5) XML coding of the control points defining the leaf sequences, couch motion, and MU; (6) delivery of the XML file in Developer Mode.
Analysis The analysis of the work here is twofold. First, the accuracy of the delivery is experimentally quantified by delivering the TMAT treatment onto phantom, and subsequently dosimetrically quantifying the agreement between the delivered and the planned dose. Second, treatment planning evaluation of the proposed technique was performed in comparison to 3D-CRT, IMRT, and en face VMAT with avoidance sectors. Delivery accuracy Verification plans were created by calculating the optimized delivery onto a solid water phantom consisting of 10 cm of solid water, a layer of film, followed by 2 cm thick solid water slab with a pinpoint PTW TX31014 ion chamber at the center, a second layer of film, and an additional 10 cm of solid water on top. The phantom was scanned using a GE Discovery PET/CT scanner. The Anisotropic Analytical Algorithm (Version 8.9.08; Varian Medical Systems, Palo Alto, CA) calculation engine was used to calculate the dose distribution. Dose distributions where compared to film (calibrated using 13 step-wedge irradiation technique). The films for detection and calibration were from the same batch, developed at the same time, and scanned using a dedicated Epson flat top scanner (Expression 10000 XL; Epson America, Inc., Long Beach, CA). Gamma and dosimetric analysis was performed by the RIT software (RIT 113 V5.3; Radiological Imaging Technology, Inc., Colorado Springs, CO). Treatment planning evaluation A treatment planning evaluation of the TMAT technique was performed on a lumpectomy patient scanned on the in-house prone breast immobilization device. Using diagnostic imaging to guide, including mammography, breast ultrasound, and breast magnetic resonance imaging, as well as the CT acquired for treatment planning, the physician contoured the lumpectomy cavity to include any surgical changes [20]. The adaptation of RTOG
0413/NSABP B-39 fractionation scheme for external beam delivery using 3D-CRT consisting of 38.50 Gy in 3.85 Gy fractions twice daily was utilized. While IMRT implementations are excluded by RTOG 0413/NSABP B-39, several institutional clinical trials have explored IMRT implementations with the similar fractionation schemes, which constitute the rationale for the use of the fractionation scheme for inversely planned deliveries (4–6). The TMAT plan was sub-sampled to 40 delivery points which were subsequently used for the optimization. Both the 3D-CRT and IMRT plans employed an identical set of 5 non-coplanar beams. The five fields had a gantry angle and couch kick of (G270°, C0°), (G90°, C5°), (G285°, C15°), (G285°, C345°), (G90°, C345°), where G indicates the gantry angle and C indicates the couch angle for each field in IEC format. An en face VMAT technique was implemented with avoidance sectors to avoid irradiation through the thoracic cavity. There were two counterclockwise VMAT en face arcs, one ranging from 75° to 130° with a couch rotation of 0°, and one from 250° to 300° with a couch rotation of 345°. All plans were optimized with exactly the same objectives and constraints. Specifically, the constraints and priorities were as follows: 95% of target volume to receive 38.5 Gy, priority 100; 0% of target volume to receive 40.5 Gy, priority 100; Varian geometrical normal tissue objective (distance from target border of 0.2 cm, start dose of 105%, end dose 60%, fall-off of 0.05), priority 150. All plans were normalized so that 100% of the prescription dose (38.5 Gy) covers 95% of the target. The same target volume with same margin was used for all treatment plans. The conformity of the plans were quantified using the RTOG conformity index, defined as CIRTOG ¼ VTVRI [21], where V RI is the volume of the reference isodose line, here taken as the 100% isodose line, and TV is the target volume. Additionally, the dose–volume parameters, V100% and V50%, that have been associated with toxicity of IMRT ABPI techniques [8,11], were determined.
Results The optimized non-collisional trajectory of the couch is presented Video 1 (CAD animation) and Video 2 (recording of dryrun) which demonstrate the available trajectory that the delivery, as formulated in the Method section, was based upon. As depicted in Fig. 2, a complex series of non-isocentric motions are required to enable for wide-angular irradiation of the target in a manner that avoids collisions. The control points to the right and left of the isocenter in the plot correspond to gantry angles of 270° and 90° IEC, respectively. The target is tracked using the MLC at the two set gantry angles, and the jaws were set to half-beam block, with the vertical coordinate of the couch set accordingly so that the photons diverge below the patient. In relation to the target, the trajectory results in approximately 210° angular distribution of beams
478
Trajectory modulated prone breast irradiation
impinging on the target for treatment planning optimization, as shown in Fig. 2c in the 3D rendering of the beam distribution about the target. The resulting sampling of the couch angles is a consequence of concurrent factors that must be satisfied in the optimization. First, any couch movement must avoid collision between the safety envelope and the LINAC. Second, the target, i.e. the lumpectomy cavity, must be maximally in the beams-eye-view of the MLC. The derived trajectory was sampled into a series of control points along which inverse planning optimization was performed. The delivery control points consisting of MLC apertures, associated monitor units, and couch positions, were combined with motion only control points (e.g., for rotating the gantry and couch to the opposing position), which were in turn programed into XML, and implemented on the LINAC as shown in Fig. 2b and exemplified in Video 2. The treatment time for the total delivery was measured to be 4 min and 56 s at a dose rate of 600 MU/min. The TMAT plan was delivered onto solid water phantom containing layers of film and a pinpoint ionization chamber to assess the accuracy of the delivery. The results of the accuracy of delivery as quantified by film dosimetry are shown in Fig. 3. Ninety three percentage of analyzed points passing the Gamma 3%/3 mm criteria, and good agreement was observed between the isodose overlay and dose profiles. The ion chamber measurement was 2.4% relative to the expected plan dose. The results of treatment planning optimization along the trajectory are presented in Fig. 4, along with corresponding planning
results using RTOG 0413 5-field conformal, 5-field IMRT, and en face VMAT with avoidance sectors. The conformity index was measured to be 1.28, 1.16, 1.37, and 1.01 for the 3D-CRT, IMRT, VMAT, and TMAT plans, respectively. The TMAT plan resulted in the lowest value of the V100% and V50% parameters. Specifically, the V100% was measured to be 6.58, 4.28, 8.85, 1.14 cm3, and the V50% was measured to be 106.37, 88.04, 77.16, 44.88 cm3 for the 3D-CRT, IMRT, VMAT, and TMAT plans, respectively. All treatment plans resulted in negligible dose to the heart and ipsilateral lung (<0.1% of the total prescription dose). Discussion The ASTRO and GEC-ESTRO recommendations for accelerated partial breast irradiation deem that APBI may be suitable for low-risk patients of older age outside of a trial, acceptable for intermediate-risk patients on a clinical trial setting, and contraindicated for high-risk patients [22,23]. While brachytherapy has been a prominent implementation of APBI, external beam techniques provide a more convenient and noninvasive alternative. However, the toxicity external beam APBI has been correlated with the conformity of the dose distribution, and specifically for IMRT implementations, to the normal tissue dose–volume parameters such as V100% and V50%. Analysis of the results indicates that the proposed trajectory modulated technique results in enhancement of the conformity, as quantified by the CI. Compared to the 5-field non-coplanar IMRT technique, which resulted in the best dosimetry of the
Fig. 2. (a) View of the optimized trajectory of the couch from the top looking down (solid black lines indicate couch trajectory as represented by the target position; dashed black lines represent motion-only control points; dashed blue lines represent the couch orientation). (b) Experimental implementation stills. (c) Beam geometry resulting from the trajectory in (a) as plotted relative to the target.
B. Fahimian et al. / Radiotherapy and Oncology 109 (2013) 475–481
479
Fig. 3. Dosimetric evaluation of TMAT delivery. (a) Gamma distribution (3%/3 mm) of the film delivery showing a pass rate of 93%. (b) Isodose overlay of the experimental detection versus the treatment planning algorithm. Comparison of vertical (c) and horizontal (d) profiles.
Fig. 4. Dosimetric comparison of (a) 3D-CRT (b) IMRT (c) VMAT (d) TMAT.
other methods considered, the V100% is reduced by 73% and the V50% is reduced by 49% for the proposed trajectory modulated
technique. It is stressed however that the treatment planning comparison here is only to serve as an illustration of the general type of
480
Trajectory modulated prone breast irradiation
dosimetry that may be achievable by such a technique – assessment of the full clinical significance will require further research using multi-patient statistical analysis of treatment planning parameters, and investigation of the clinically relevant end-points. While such a study is beyond the scope of this initial development and implementation paper, the reported improved dosimetric results suggest such an approach is warranted, and will be the subject of future work. In a simulation paper, the potential for continuous motion of the couch for a supine breast setup was proposed [19]. Here, a prone setup was specifically utilized over a supine setup for several reasons. For appropriately selected patients, placement in the prone position provides maximal separation of the lumpectomy cavity from the thorax, and allows greater degrees of freedom in the choice of beam irradiation geometry [24]. Additionally, the prone setup effectively negates respiratory motion by stabilizing the patient on the chest and forcing breathing motion to be transferred to the back. However, in the supine method, respiratory motion management may not be fully addressed, and expansion of the treatment volume or gating of the couch motion may need to be implemented. With respect to the initial approach in [19], several other key developments are presented. First, instead of a uniform continuous rotation of the couch, here a robust trajectory optimization method is developed from which it is shown that a complex non-uniform and non-isocentric trajectory as in Fig. 2a is required for wide-angular irradiation. Second, relative to [19] which utilized conformal irradiation with limited rotation of the couch (resulting in limited dosimetric advantage over static non-coplanar IMRT), the approach here builds upon this work by integrating inversely planned intensity modulated delivery with the motion of the couch, and results in significant dosimetric advantage over static non-coplanar IMRT. Finally, the work here presents the first full experimental implementation and dosimetric validation of dynamic breast irradiation using motion of the couch. As with all techniques, appropriate patient selection is however implied. For the patients who may be suitable for a prone setup, this technique may allow for improved dosimetry. It is noted that in the technique implemented in this work, the gantry angles were restricted to two static angles, with the target being tracked by the MLC. If consistent immobilization is used for different patients such that the targets approximately fall in the same region, the derived trajectories in this work may provide a general template or initial starting trajectory for the optimization. The method can be readily generalized to include tilting or dynamic motion of the gantry, which may present applications for irradiating targets close to the thoracic wall – such an extension will be the subject of future work.
Conclusion A LINAC based approach for partial breast irradiation has been developed which through coordinated motion of the couch, intensity modulated delivery, and a prone breast setup enables robust wide angle modulated irradiation of the ipsilateral breast in a manner that avoids irradiation through the thorax and contralateral breast. The feasibility of such a technique combining motion of the couch and intensity modulated delivery has been demonstrated experimentally on a standard LINAC for the first time, and the accuracy of the delivery has been validated. Treatment planning results indicate that this methodology provides for enhancement of dose conformity and associated dose–volume parameters to minimize toxicity compared to other APBI techniques. For the population of patients eligible for a prone breast setup, this technique enables a LINAC based solution for highly conformal treatments with more adequate sparing normal tissue.
Conflict of interest None. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.radonc.2013.10. 031. References [1] Clark RM, McCulloch PB, Levine MN, Lipa M, Wilkinson RH, Mahoney LJ, et al. Randomized clinical trial to assess the effectiveness of breast irradiation following lumpectomy and axillary dissection for node-negative breast cancer. J Natl Cancer Inst 1992;84:683–9. [2] Liljegren G, Holmberg L, Bergh J, Lindgren A, Tabar L, Nordgren H, et al. 10-Year results after sector resection with or without postoperative radiotherapy for stage I breast cancer: a randomized trial. J Clin Oncol 1999;17:2326–33. [3] Veronesi U, Cascinelli N, Mariani L, Greco M, Saccozzi R, Luini A, et al. Twentyyear follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med 2002;347: 1227–32. [4] Offersen BV, Overgaard M, Kroman N, Overgaard J. Accelerated partial breast irradiation as part of breast conserving therapy of early breast carcinoma: a systematic review. Radiother Oncol 2009;90:1–13. [5] Mannino M, Yarnold J. Accelerated partial breast irradiation trials: diversity in rationale and design. Radiother Oncol 2009;91:16–22. [6] Mukesh M, Harris E, Jena R, Evans P, Coles C. Relationship between irradiated breast volume and late normal tissue complications: a systematic review. Radiother Oncol 2012;104:1–10. [7] Baglan KL, Sharpe MB, Jaffray D, Frazier RC, Fayad J, Kestin LL, et al. Accelerated partial breast irradiation using 3D conformal radiation therapy (3D-CRT). Int J Radiat Oncol Biol Phys 2003;55:302–11. [8] Jagsi R, Ben-David MA, Moran JM, Marsh RB, Griffith KA, Hayman JA, et al. Unacceptable cosmesis in a protocol investigating intensity-modulated radiotherapy with active breathing control for accelerated partial-breast irradiation. Int J Radiat Oncol Biol Phys 2010;76:71–8. [9] Lewin AA, Derhagopian R, Saigal K, Panoff JE, Abitbol A, Wieczorek DJ, et al. Accelerated partial breast irradiation is safe and effective using intensitymodulated radiation therapy in selected early-stage breast cancer. Int J Radiat Oncol Biol Phys 2012;82:2104–10. [10] Livi L, Buonamici FB, Simontacchi G, Scotti V, Fambrini M, Compagnucci A, et al. Accelerated partial breast irradiation with IMRT: new technical approach and interim analysis of acute toxicity in a phase III randomized clinical trial. Int J Radiat Oncol Biol Phys 2010;77:509–15. [11] Leonard KL, Hepel JT, Hiatt JR, Dipetrillo TA, Price LL, Wazer DE. The effect of dose–volume parameters and interfraction interval on cosmetic outcome and toxicity after 3-dimensional conformal accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys 2013;85:623–9. [12] Smith GL, Xu Y, Buchholz TA, Giordano SH, Jiang J, Shih YC, et al. Association between treatment with brachytherapy vs whole-breast irradiation and subsequent mastectomy, complications, and survival among older women with invasive breast cancer. JAMA 2012;307:1827–37. [13] Hepel JT, Tokita M, MacAusland SG, Evans SB, Hiatt JR, Price LL, et al. Toxicity of three-dimensional conformal radiotherapy for accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys 2009;75:1290–6. [14] Feigenberg SJ, Yu CX, Regine WF. Development of a stereotactic body radiation therapy (SBRT) system dedicated to the treatment of early stage breast cancer: the Gammapod. Int J Radiat Oncol Biol Phys 2010;78:S208. [15] Vermeulen S, Cotrutz C, Morris A, Meier R, Buchanan C, Dawson P, et al. Accelerated partial breast irradiation: using the CyberKnife as the radiation delivery platform in the treatment of early breast cancer. Front Oncol 2011;1:43. [16] Wang X, Amos RA, Zhang X, Taddei PJ, Woodward WA, Hoffman KE, et al. External-beam accelerated partial breast irradiation using multiple proton beam configurations. Int J Radiat Oncol Biol Phys 2011;80:1464–72. [17] Bush DA, Slater JD, Garberoglio C, Yuh G, Hocko JM, Slater JM. A technique of partial breast irradiation utilizing proton beam radiotherapy: comparison with conformal X-ray therapy. Cancer J 2007;13:114–8. [18] Moon SH, Shin KH, Kim TH, Yoon M, Park S, Lee D-H, et al. Dosimetric comparison of four different external beam partial breast irradiation techniques: three-dimensional conformal radiotherapy, intensity-modulated radiotherapy, helical tomotherapy, and proton beam therapy. Radiother Oncol 2009;90:66–73. [19] Shaitelman SF, Kim LH, Yan D, Martinez AA, Vicini FA, Grills IS. Continuous arc rotation of the couch therapy for the delivery of accelerated partial breast irradiation: a treatment planning analysis. Int J Radiat Oncol Biol Phys 2011;80:771–8. [20] Yang TJ, Tao R, Elkhuizen PHM, van Vliet-Vroegindeweij C, Li G, Powell SN. Tumor bed delineation for external beam accelerated partial breast irradiation: a systematic review. Radiother Oncol 2013;108:181–9.
B. Fahimian et al. / Radiotherapy and Oncology 109 (2013) 475–481 [21] Feuvret L, Noel G, Mazeron JJ, Bey P. Conformity index: a review. Int J Radiat Oncol Biol Phys 2006;64:333–42. [22] Smith BD, Arthur DW, Buchholz TA, Haffty BG, Hahn CA, Hardenbergh PH, et al. Accelerated partial breast irradiation consensus statement from the American Society for Radiation Oncology (ASTRO). Int J Radiat Oncol Biol Phys 2009; 74:987–1001. [23] Polgar C, Van Limbergen E, Potter R, Kovacs G, Polo A, Lyczek J, et al. Patient selection for accelerated partial-breast irradiation (APBI) after breast-
481
conserving surgery: recommendations of the Groupe Europeen de Curietherapie-European Society for Therapeutic Radiology and Oncology (GEC-ESTRO) breast cancer working group based on clinical evidence (2009). Radiother Oncol 2010;94:264–73. [24] Becker SJ, Elliston C, Dewyngaert K, Jozsef G, Brenner D, Formenti S. Breast radiotherapy in the prone position primarily reduces the maximum out-offield measured dose to the ipsilateral lung. Med Phys 2012;39:2417–23.
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Trajectory modulated arc therapy
Trajectory modulated prone breast irradiation: A LINAC-based technique combining intensity modulated delivery and motion of the couch Benjamin Fahimian, Victoria Yu, Kathleen Horst, Lei Xing, Dimitre Hristov ⇑ Department of Radiation Oncology, Stanford University, United States
a r t i c l e
i n f o
Article history: Received 14 December 2012 Received in revised form 16 October 2013 Accepted 19 October 2013 Available online 11 November 2013 Keywords: Accelerated partial breast irradiation (APBI) Couch motion IMRT Trajectory modulated arc therapy (TMAT)
a b s t r a c t Purpose: External beam radiation therapy (EBRT) provides a non-invasive treatment alternative for accelerated partial breast irradiation (APBI), however, limitations in achievable dose conformity of current EBRT techniques have been correlated to reported toxicity. To enhance the conformity of EBRT APBI, a technique for conventional LINACs is developed, which through combined motion of the couch, intensity modulated delivery, and a prone breast setup, enables wide-angular coronal arc irradiation of the ipsilateral breast without irradiating through the thorax and contralateral breast. Methods and materials: A couch trajectory optimization technique was developed to determine the trajectories that concurrently avoid collision with the LINAC and maintain the target within the MLC apertures. Inverse treatment planning was performed along the derived trajectory. The technique was experimentally implemented by programming the Varian TrueBeam™ STx in Developer Mode. The dosimetric accuracy of the delivery was evaluated by ion chamber and film measurements in phantom. Results: The resulting optimized trajectory was shown to be necessarily non-isocentric, and contain both translation and rotations of the couch. Film measurements resulted in 93% of the points in the measured two-dimensional dose maps passing the 3%/3 mm Gamma criterion. Preliminary treatment plan comparison to 5-field 3D-conformal, IMRT, and VMAT demonstrated enhancement in conformity, and reduction of the normal tissue V50% and V100% parameters that have been correlated with EBRT toxicity. Conclusions: The feasibility of wide-angular intensity modulated partial breast irradiation using motion of the couch has been demonstrated experimentally on a standard LINAC for the first time. For patients eligible for a prone setup, the technique may enable improvement of dose conformity and associated dose–volume parameters correlated with toxicity. Ó 2013 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 109 (2013) 475–481
Breast conservation therapy consisting of lumpectomy followed by adjuvant radiation therapy is an established treatment for early stage breast cancer. While whole breast irradiation (WBI) is the technique of proven efficacy, alternative techniques using partial breast irradiation of the lumpectomy cavity are under active investigation based on data suggesting that the majority of in breast recurrences occur in the vicinity of the lumpectomy cavity [1–3]. These accelerated partial breast irradiation (APBI) regimens, which additionally allow for a reduction of treatment time to 1 week versus the traditional 6 weeks of WBI, have garnered increased clinical implementation in recent years [4]. Data on the long term efficacy and toxicity of such APBI techniques are still emerging, and ongoing studies such as the Radiation Therapy Oncology Group (RTOG) 0413/National Surgical Adjuvant Breast and Bowel Project (NSABP) B-39 randomized Phase III trial will provide data comparing WBI and APBI [5]. Unlike WBI which aims to uniformly irradiate the breast, partial breast irradiation attempts to conformally irradiate ⇑ Corresponding author. Address: 875 Blake Wilbur Drive, G203, Stanford, CA 94305-5487, United States. E-mail address: [email protected] (D. Hristov). 0167-8140/$ - see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.radonc.2013.10.031
the lumpectomy cavity plus a margin while minimizing dose to the surrounding normal tissue. Since different radiation delivery modalities result in different conformity and normal tissue dose, it is noted that the toxicity of APBI is additionally dependent on the radiation technique [6]. External beam radiotherapy (EBRT) techniques using 3D conformal radiation therapy (3D-CRT) [7] and intensity modulated radiation therapy (IMRT) [8–11] are attractive options for partial breast irradiation because these techniques are non-invasive, widely available and simpler to implement in comparison to brachytherapy alternatives [12]. However, with existing EBRT APBI approaches, unacceptably high rates of moderate-to-severe late normal tissue toxicities have been correlated with several dose– volume parameters [8,11,13]. Specifically, the occurrence of unacceptable cosmesis was correlated with the V100% and V50% [8,11]. Such toxicity can be partially attributed to the fact that current techniques are limited in the achievable dose conformity as they deploy a limited number of tangential field arrangements, which while sparing thoracic normal tissue, inherently irradiate large volumes of breast to moderate and high doses due to limitations in degrees of freedom in the beam arrangements.
476
Trajectory modulated prone breast irradiation
Several alternative methods have been proposed for external beam APBI. These include dedicated stereotactic 60Co breast systems [14], robotic radiosurgery [15], and proton beams [16–18]. While these techniques may allow for variable degrees of improved dosimetry, they require specialized or dedicated systems, which are typically outside the realm of the capabilities of most clinics, and may be questionable in the scope of cost-effective healthcare management. It is thus desirable to develop more conformal types of delivery techniques under conventional LINAC geometry. As such, others have proposed the utilization of a uniform couch rotation under a supine setup [19]. However, no significant dosimetric advantage was achieved relative to a 5-field IMRT technique due to the fact that only a limited uniform continuous couch rotation was modeled (please refer to discussion section for further details). To address the complex dosimetric challenge that partial breast irradiation presents, in this work a delivery method for a conventional LINAC is introduced and developed which combines a nonuniform and non-isocentric couch trajectory, inversely planned intensity modulated delivery, and prone breast setup to enable highly conformal treatments through intensity modulated coronal arc irradiation of the lumpectomy cavity, in a manner avoiding irradiation through the thorax. Specifically, using a method of trajectory based optimization the optimal couch trajectory that concurrently provides wide-angular coverage, avoids collision between the couch and LINAC components, and maintains the target in the beams-eye-view (BEV) of the multi-leaf collimator (MLC), is determined. Inverse planning is used to design an intensity modulated delivery about the trajectory. The feasibility of the proposed trajectory modulated arc therapy (TMAT) delivery technique is experimentally quantified by programming the trajectory modulated delivery on a conventional LINAC. The accuracy of the delivery technique is then quantified using film and ion chamber measurements. While full analysis of multi-patient treatment planning comparisons is beyond the scope of this initial implementation and feasibility work, to demonstrate the type of dosimetry that may be achievable through the technique, a preliminary planning comparison is provided for the delivered case comparing dosimetry of respective implementations using 3D-CRT, IMRT, and en face volumetric modulated arc therapy (VMAT) with avoidance sectors. Method Trajectory optimization The optimization of the delivery trajectory incorporating the motion of the couch entails the determination of the synchronized couch and gantry trajectory which concurrently (1) avoids collisions between all components of the LINAC and the patient, (2) keeps the target in the BEV of the MLC, and (3) provides maximal angular spread of beams. A 4D computer aided design (CAD) model of the LINAC and couch, enabling visualization of the BEV relative to imported patient scans and detection of collisions, was established in SolidWorks™ (Dassault Systèmes SolidWorks Corp., Massachusetts, USA). The trajectory optimization was formulated with the couch modeled as a rigid object, defined as subset of a set of points C in three-dimensional space, C 2 R3 . The transformation of points due to three translational axes and one rotational axis is denoted by combined time dependent transformation Tt ðCÞ : R3 ! R3 . It is noted that the set of points denoted as C includes the patient volume, immobilization device, and target, all encompassed in a safety envelope, as shown in Fig. 1. The LINAC and room geometry is given by a configuration in space denoted by L 2 R3 , which depends on the specific time-dependent orientation of the LINAC.
Subsequently, under a non-collisional couch and LINAC state, the intersection of the two sets is the null set, i.e., C \ L ¼ ;, while a collision is detected if C \ L–;. Besides a non-collisional state between the couch and LINAC components, the trajectory optimization requires that the target be in the field of view of the MLC. A projectional operator is defined as PðMLCÞ : R3 ! R2 and PðPTVÞ : R3 ! R2 , which, at the specific couch angle, projects either the outline of the MLCs or the PTV in a plane orthogonal to the central axis of the beam and passing through the isocenter. This operation yields the point sets of the MLC and PTV projections denoted as P MLC 2 R2 and PPTV 2 R2 , respectively. For a valid beam delivery control point it is required that P MLC \ P PTV –;, that is at least some portion of the PTV should be within the BEV defined by the MLC. The pseudo-code for the trajectory optimization is given in the algorithm in Table 1. All angles are in International Electrotechnical Commission (IEC) 61217 format.
Experimental implementation Currently, no commercial LINAC allows for synchronized motion of the couch during clinical delivery. In order to implement the technique, the Varian TrueBeam™ STx (Serial # 3; Varian Medical Systems, Palo Alto, CA) was utilized in conjunction with Developer Mode capabilities. Under a research environment, Developer Mode allows for custom XML scripting and control of all components of the system, through which, the optimized parameters for motion of the couch, MLC, and beam delivery were programed in this work. The LINAC configuration includes the HD120 MLC, which consists of 60 pairs of leaves, with the central 32 pairs of width of 0.25 cm, flogged symmetrically by 26 pairs of 0.50 cm, and a pair of 0.78 cm MLCs at each edge, resulting in an approximate effective coverage at isocenter of 22 36 cm (18.1 cm maximum extension beyond collimator x-jaw). The maximum supported leaf speed is 2.5 cm/s, and the average leaf transmission averaged across all leaves is <2%. For all implementations in this work, the flattened 6 MV X-ray beam at a dose-rate of 600 MU/min was utilized. The treatment platform included an in-house breast board (enabling elevation of the patient body above the breast opening), coupled with the 4D Exact™ IGRT carbon fiber couch (three degrees of freedom in translation and one in rotation). The Varian Developer Mode schema defines the system control via a Trajectory Model in which all mechanical axis positions and beam delivery parameters are specified with associated trajectory functions. The trajectory functions are defined by the user at discrete cumulative MU control points, in between which a linear
Fig. 1. 4D CAD model of the LINAC, couch, target, and immobilization device utilized for the trajectory optimization.
B. Fahimian et al. / Radiotherapy and Oncology 109 (2013) 475–481
477
Table 1 Trajectory generation algorithm. Prone APBI trajectory generation { SET Gantry ROT = 90°; Couch ROT = 270°; Couch LAT = limit; Couch LONG = limit; Couch VERT such that P MLC \ P PTV –;; SUB-ALGORITHM 1{ IF C \ L–; // No collision between couch or LINAC, // But may not be useful if target outside BEV or MLC limits IF P MLC \ P PTV –; //Target is in beam’s eye view WRITE //Record as delivery control point; END END} Tt ðCÞ : R3 ! R3 ;
// Increment couch either because control point found, // collision detected, or target outside BEV GOTO SUB-ALGORITHM 1 WHILE Couch LONG and ROT < Limit REPEAT for Gantry ROT = 270°}
transformation in the positions and states is interpreted by the system. The experimental implementation of the method proposed here entailed: (1) development of a time-dependent CAD model of the LINAC environment; (2) import of patient CT and contours into the model; (3) optimization of the non-collisional lateral and medial couch trajectories by the algorithm in Table 1, followed by 4D CAD emulation verification; (4) treatment planning optimization by Varian Dose Volume Optimizer, followed by leaf sequencing of the IMRT fluence maps into up to 10 deliverable MLC segments per field; (5) XML coding of the control points defining the leaf sequences, couch motion, and MU; (6) delivery of the XML file in Developer Mode.
Analysis The analysis of the work here is twofold. First, the accuracy of the delivery is experimentally quantified by delivering the TMAT treatment onto phantom, and subsequently dosimetrically quantifying the agreement between the delivered and the planned dose. Second, treatment planning evaluation of the proposed technique was performed in comparison to 3D-CRT, IMRT, and en face VMAT with avoidance sectors. Delivery accuracy Verification plans were created by calculating the optimized delivery onto a solid water phantom consisting of 10 cm of solid water, a layer of film, followed by 2 cm thick solid water slab with a pinpoint PTW TX31014 ion chamber at the center, a second layer of film, and an additional 10 cm of solid water on top. The phantom was scanned using a GE Discovery PET/CT scanner. The Anisotropic Analytical Algorithm (Version 8.9.08; Varian Medical Systems, Palo Alto, CA) calculation engine was used to calculate the dose distribution. Dose distributions where compared to film (calibrated using 13 step-wedge irradiation technique). The films for detection and calibration were from the same batch, developed at the same time, and scanned using a dedicated Epson flat top scanner (Expression 10000 XL; Epson America, Inc., Long Beach, CA). Gamma and dosimetric analysis was performed by the RIT software (RIT 113 V5.3; Radiological Imaging Technology, Inc., Colorado Springs, CO). Treatment planning evaluation A treatment planning evaluation of the TMAT technique was performed on a lumpectomy patient scanned on the in-house prone breast immobilization device. Using diagnostic imaging to guide, including mammography, breast ultrasound, and breast magnetic resonance imaging, as well as the CT acquired for treatment planning, the physician contoured the lumpectomy cavity to include any surgical changes [20]. The adaptation of RTOG
0413/NSABP B-39 fractionation scheme for external beam delivery using 3D-CRT consisting of 38.50 Gy in 3.85 Gy fractions twice daily was utilized. While IMRT implementations are excluded by RTOG 0413/NSABP B-39, several institutional clinical trials have explored IMRT implementations with the similar fractionation schemes, which constitute the rationale for the use of the fractionation scheme for inversely planned deliveries (4–6). The TMAT plan was sub-sampled to 40 delivery points which were subsequently used for the optimization. Both the 3D-CRT and IMRT plans employed an identical set of 5 non-coplanar beams. The five fields had a gantry angle and couch kick of (G270°, C0°), (G90°, C5°), (G285°, C15°), (G285°, C345°), (G90°, C345°), where G indicates the gantry angle and C indicates the couch angle for each field in IEC format. An en face VMAT technique was implemented with avoidance sectors to avoid irradiation through the thoracic cavity. There were two counterclockwise VMAT en face arcs, one ranging from 75° to 130° with a couch rotation of 0°, and one from 250° to 300° with a couch rotation of 345°. All plans were optimized with exactly the same objectives and constraints. Specifically, the constraints and priorities were as follows: 95% of target volume to receive 38.5 Gy, priority 100; 0% of target volume to receive 40.5 Gy, priority 100; Varian geometrical normal tissue objective (distance from target border of 0.2 cm, start dose of 105%, end dose 60%, fall-off of 0.05), priority 150. All plans were normalized so that 100% of the prescription dose (38.5 Gy) covers 95% of the target. The same target volume with same margin was used for all treatment plans. The conformity of the plans were quantified using the RTOG conformity index, defined as CIRTOG ¼ VTVRI [21], where V RI is the volume of the reference isodose line, here taken as the 100% isodose line, and TV is the target volume. Additionally, the dose–volume parameters, V100% and V50%, that have been associated with toxicity of IMRT ABPI techniques [8,11], were determined.
Results The optimized non-collisional trajectory of the couch is presented Video 1 (CAD animation) and Video 2 (recording of dryrun) which demonstrate the available trajectory that the delivery, as formulated in the Method section, was based upon. As depicted in Fig. 2, a complex series of non-isocentric motions are required to enable for wide-angular irradiation of the target in a manner that avoids collisions. The control points to the right and left of the isocenter in the plot correspond to gantry angles of 270° and 90° IEC, respectively. The target is tracked using the MLC at the two set gantry angles, and the jaws were set to half-beam block, with the vertical coordinate of the couch set accordingly so that the photons diverge below the patient. In relation to the target, the trajectory results in approximately 210° angular distribution of beams
478
Trajectory modulated prone breast irradiation
impinging on the target for treatment planning optimization, as shown in Fig. 2c in the 3D rendering of the beam distribution about the target. The resulting sampling of the couch angles is a consequence of concurrent factors that must be satisfied in the optimization. First, any couch movement must avoid collision between the safety envelope and the LINAC. Second, the target, i.e. the lumpectomy cavity, must be maximally in the beams-eye-view of the MLC. The derived trajectory was sampled into a series of control points along which inverse planning optimization was performed. The delivery control points consisting of MLC apertures, associated monitor units, and couch positions, were combined with motion only control points (e.g., for rotating the gantry and couch to the opposing position), which were in turn programed into XML, and implemented on the LINAC as shown in Fig. 2b and exemplified in Video 2. The treatment time for the total delivery was measured to be 4 min and 56 s at a dose rate of 600 MU/min. The TMAT plan was delivered onto solid water phantom containing layers of film and a pinpoint ionization chamber to assess the accuracy of the delivery. The results of the accuracy of delivery as quantified by film dosimetry are shown in Fig. 3. Ninety three percentage of analyzed points passing the Gamma 3%/3 mm criteria, and good agreement was observed between the isodose overlay and dose profiles. The ion chamber measurement was 2.4% relative to the expected plan dose. The results of treatment planning optimization along the trajectory are presented in Fig. 4, along with corresponding planning
results using RTOG 0413 5-field conformal, 5-field IMRT, and en face VMAT with avoidance sectors. The conformity index was measured to be 1.28, 1.16, 1.37, and 1.01 for the 3D-CRT, IMRT, VMAT, and TMAT plans, respectively. The TMAT plan resulted in the lowest value of the V100% and V50% parameters. Specifically, the V100% was measured to be 6.58, 4.28, 8.85, 1.14 cm3, and the V50% was measured to be 106.37, 88.04, 77.16, 44.88 cm3 for the 3D-CRT, IMRT, VMAT, and TMAT plans, respectively. All treatment plans resulted in negligible dose to the heart and ipsilateral lung (<0.1% of the total prescription dose). Discussion The ASTRO and GEC-ESTRO recommendations for accelerated partial breast irradiation deem that APBI may be suitable for low-risk patients of older age outside of a trial, acceptable for intermediate-risk patients on a clinical trial setting, and contraindicated for high-risk patients [22,23]. While brachytherapy has been a prominent implementation of APBI, external beam techniques provide a more convenient and noninvasive alternative. However, the toxicity external beam APBI has been correlated with the conformity of the dose distribution, and specifically for IMRT implementations, to the normal tissue dose–volume parameters such as V100% and V50%. Analysis of the results indicates that the proposed trajectory modulated technique results in enhancement of the conformity, as quantified by the CI. Compared to the 5-field non-coplanar IMRT technique, which resulted in the best dosimetry of the
Fig. 2. (a) View of the optimized trajectory of the couch from the top looking down (solid black lines indicate couch trajectory as represented by the target position; dashed black lines represent motion-only control points; dashed blue lines represent the couch orientation). (b) Experimental implementation stills. (c) Beam geometry resulting from the trajectory in (a) as plotted relative to the target.
B. Fahimian et al. / Radiotherapy and Oncology 109 (2013) 475–481
479
Fig. 3. Dosimetric evaluation of TMAT delivery. (a) Gamma distribution (3%/3 mm) of the film delivery showing a pass rate of 93%. (b) Isodose overlay of the experimental detection versus the treatment planning algorithm. Comparison of vertical (c) and horizontal (d) profiles.
Fig. 4. Dosimetric comparison of (a) 3D-CRT (b) IMRT (c) VMAT (d) TMAT.
other methods considered, the V100% is reduced by 73% and the V50% is reduced by 49% for the proposed trajectory modulated
technique. It is stressed however that the treatment planning comparison here is only to serve as an illustration of the general type of
480
Trajectory modulated prone breast irradiation
dosimetry that may be achievable by such a technique – assessment of the full clinical significance will require further research using multi-patient statistical analysis of treatment planning parameters, and investigation of the clinically relevant end-points. While such a study is beyond the scope of this initial development and implementation paper, the reported improved dosimetric results suggest such an approach is warranted, and will be the subject of future work. In a simulation paper, the potential for continuous motion of the couch for a supine breast setup was proposed [19]. Here, a prone setup was specifically utilized over a supine setup for several reasons. For appropriately selected patients, placement in the prone position provides maximal separation of the lumpectomy cavity from the thorax, and allows greater degrees of freedom in the choice of beam irradiation geometry [24]. Additionally, the prone setup effectively negates respiratory motion by stabilizing the patient on the chest and forcing breathing motion to be transferred to the back. However, in the supine method, respiratory motion management may not be fully addressed, and expansion of the treatment volume or gating of the couch motion may need to be implemented. With respect to the initial approach in [19], several other key developments are presented. First, instead of a uniform continuous rotation of the couch, here a robust trajectory optimization method is developed from which it is shown that a complex non-uniform and non-isocentric trajectory as in Fig. 2a is required for wide-angular irradiation. Second, relative to [19] which utilized conformal irradiation with limited rotation of the couch (resulting in limited dosimetric advantage over static non-coplanar IMRT), the approach here builds upon this work by integrating inversely planned intensity modulated delivery with the motion of the couch, and results in significant dosimetric advantage over static non-coplanar IMRT. Finally, the work here presents the first full experimental implementation and dosimetric validation of dynamic breast irradiation using motion of the couch. As with all techniques, appropriate patient selection is however implied. For the patients who may be suitable for a prone setup, this technique may allow for improved dosimetry. It is noted that in the technique implemented in this work, the gantry angles were restricted to two static angles, with the target being tracked by the MLC. If consistent immobilization is used for different patients such that the targets approximately fall in the same region, the derived trajectories in this work may provide a general template or initial starting trajectory for the optimization. The method can be readily generalized to include tilting or dynamic motion of the gantry, which may present applications for irradiating targets close to the thoracic wall – such an extension will be the subject of future work.
Conclusion A LINAC based approach for partial breast irradiation has been developed which through coordinated motion of the couch, intensity modulated delivery, and a prone breast setup enables robust wide angle modulated irradiation of the ipsilateral breast in a manner that avoids irradiation through the thorax and contralateral breast. The feasibility of such a technique combining motion of the couch and intensity modulated delivery has been demonstrated experimentally on a standard LINAC for the first time, and the accuracy of the delivery has been validated. Treatment planning results indicate that this methodology provides for enhancement of dose conformity and associated dose–volume parameters to minimize toxicity compared to other APBI techniques. For the population of patients eligible for a prone breast setup, this technique enables a LINAC based solution for highly conformal treatments with more adequate sparing normal tissue.
Conflict of interest None. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.radonc.2013.10. 031. References [1] Clark RM, McCulloch PB, Levine MN, Lipa M, Wilkinson RH, Mahoney LJ, et al. Randomized clinical trial to assess the effectiveness of breast irradiation following lumpectomy and axillary dissection for node-negative breast cancer. J Natl Cancer Inst 1992;84:683–9. [2] Liljegren G, Holmberg L, Bergh J, Lindgren A, Tabar L, Nordgren H, et al. 10-Year results after sector resection with or without postoperative radiotherapy for stage I breast cancer: a randomized trial. J Clin Oncol 1999;17:2326–33. [3] Veronesi U, Cascinelli N, Mariani L, Greco M, Saccozzi R, Luini A, et al. Twentyyear follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med 2002;347: 1227–32. [4] Offersen BV, Overgaard M, Kroman N, Overgaard J. Accelerated partial breast irradiation as part of breast conserving therapy of early breast carcinoma: a systematic review. Radiother Oncol 2009;90:1–13. [5] Mannino M, Yarnold J. Accelerated partial breast irradiation trials: diversity in rationale and design. Radiother Oncol 2009;91:16–22. [6] Mukesh M, Harris E, Jena R, Evans P, Coles C. Relationship between irradiated breast volume and late normal tissue complications: a systematic review. Radiother Oncol 2012;104:1–10. [7] Baglan KL, Sharpe MB, Jaffray D, Frazier RC, Fayad J, Kestin LL, et al. Accelerated partial breast irradiation using 3D conformal radiation therapy (3D-CRT). Int J Radiat Oncol Biol Phys 2003;55:302–11. [8] Jagsi R, Ben-David MA, Moran JM, Marsh RB, Griffith KA, Hayman JA, et al. Unacceptable cosmesis in a protocol investigating intensity-modulated radiotherapy with active breathing control for accelerated partial-breast irradiation. Int J Radiat Oncol Biol Phys 2010;76:71–8. [9] Lewin AA, Derhagopian R, Saigal K, Panoff JE, Abitbol A, Wieczorek DJ, et al. Accelerated partial breast irradiation is safe and effective using intensitymodulated radiation therapy in selected early-stage breast cancer. Int J Radiat Oncol Biol Phys 2012;82:2104–10. [10] Livi L, Buonamici FB, Simontacchi G, Scotti V, Fambrini M, Compagnucci A, et al. Accelerated partial breast irradiation with IMRT: new technical approach and interim analysis of acute toxicity in a phase III randomized clinical trial. Int J Radiat Oncol Biol Phys 2010;77:509–15. [11] Leonard KL, Hepel JT, Hiatt JR, Dipetrillo TA, Price LL, Wazer DE. The effect of dose–volume parameters and interfraction interval on cosmetic outcome and toxicity after 3-dimensional conformal accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys 2013;85:623–9. [12] Smith GL, Xu Y, Buchholz TA, Giordano SH, Jiang J, Shih YC, et al. Association between treatment with brachytherapy vs whole-breast irradiation and subsequent mastectomy, complications, and survival among older women with invasive breast cancer. JAMA 2012;307:1827–37. [13] Hepel JT, Tokita M, MacAusland SG, Evans SB, Hiatt JR, Price LL, et al. Toxicity of three-dimensional conformal radiotherapy for accelerated partial breast irradiation. Int J Radiat Oncol Biol Phys 2009;75:1290–6. [14] Feigenberg SJ, Yu CX, Regine WF. Development of a stereotactic body radiation therapy (SBRT) system dedicated to the treatment of early stage breast cancer: the Gammapod. Int J Radiat Oncol Biol Phys 2010;78:S208. [15] Vermeulen S, Cotrutz C, Morris A, Meier R, Buchanan C, Dawson P, et al. Accelerated partial breast irradiation: using the CyberKnife as the radiation delivery platform in the treatment of early breast cancer. Front Oncol 2011;1:43. [16] Wang X, Amos RA, Zhang X, Taddei PJ, Woodward WA, Hoffman KE, et al. External-beam accelerated partial breast irradiation using multiple proton beam configurations. Int J Radiat Oncol Biol Phys 2011;80:1464–72. [17] Bush DA, Slater JD, Garberoglio C, Yuh G, Hocko JM, Slater JM. A technique of partial breast irradiation utilizing proton beam radiotherapy: comparison with conformal X-ray therapy. Cancer J 2007;13:114–8. [18] Moon SH, Shin KH, Kim TH, Yoon M, Park S, Lee D-H, et al. Dosimetric comparison of four different external beam partial breast irradiation techniques: three-dimensional conformal radiotherapy, intensity-modulated radiotherapy, helical tomotherapy, and proton beam therapy. Radiother Oncol 2009;90:66–73. [19] Shaitelman SF, Kim LH, Yan D, Martinez AA, Vicini FA, Grills IS. Continuous arc rotation of the couch therapy for the delivery of accelerated partial breast irradiation: a treatment planning analysis. Int J Radiat Oncol Biol Phys 2011;80:771–8. [20] Yang TJ, Tao R, Elkhuizen PHM, van Vliet-Vroegindeweij C, Li G, Powell SN. Tumor bed delineation for external beam accelerated partial breast irradiation: a systematic review. Radiother Oncol 2013;108:181–9.
B. Fahimian et al. / Radiotherapy and Oncology 109 (2013) 475–481 [21] Feuvret L, Noel G, Mazeron JJ, Bey P. Conformity index: a review. Int J Radiat Oncol Biol Phys 2006;64:333–42. [22] Smith BD, Arthur DW, Buchholz TA, Haffty BG, Hahn CA, Hardenbergh PH, et al. Accelerated partial breast irradiation consensus statement from the American Society for Radiation Oncology (ASTRO). Int J Radiat Oncol Biol Phys 2009; 74:987–1001. [23] Polgar C, Van Limbergen E, Potter R, Kovacs G, Polo A, Lyczek J, et al. Patient selection for accelerated partial-breast irradiation (APBI) after breast-
481
conserving surgery: recommendations of the Groupe Europeen de Curietherapie-European Society for Therapeutic Radiology and Oncology (GEC-ESTRO) breast cancer working group based on clinical evidence (2009). Radiother Oncol 2010;94:264–73. [24] Becker SJ, Elliston C, Dewyngaert K, Jozsef G, Brenner D, Formenti S. Breast radiotherapy in the prone position primarily reduces the maximum out-offield measured dose to the ipsilateral lung. Med Phys 2012;39:2417–23.