A hybrid planning strategy for stereotactic body radiation therapy of early stage non‐small‐cell lung cancer

Abstract Currently dynamic conformal arcs (DCA) and volumetric modulated arc therapy (VMAT) are two popular planning techniques to treat lung stereotactic body radiation therapy (SBRT) patients. Of the two, DCA has advantages in terms of multi‐leaf collimator (MLC) motion, positioning error, and delivery efficiency. However, VMAT is often the choice when critical organ sparing becomes important. We developed a hybrid strategy to incorporate DCA component into VMAT planning, results were compared with DCA and VMAT plans. Four planning techniques were retrospectively simulated for 10 lung SBRT patients: DCA, Hybrid‐DCA (2/3 of the doses from DCA beams), Hybrid‐VMAT (2/3 of the doses from VMAT beams) and VMAT. Plan complexity was accessed by modulation complexity score (MCS). Conformity index (CI) for the planning target volume (PTV), V 20 and V 5 for the lung, V 30 for the chestwall, and maximum dose to all other critical organs were calculated. Plans were compared with regard to these metrics and measured agreement between the planned and delivered doses. DCA technique did not result in acceptable plan quality due to target location for five patients. Hybrid‐DCA produced one unacceptable plan, and Hybrid‐VMAT and VMAT produced no unacceptable plans. The CI improved with increasing VMAT usage, as did the dose sparing to critical structures. Compared to the VMAT technique, a total MU reduction of 14%, 25% and 37% were found for Hybrid‐VMAT, Hybrid‐DCA and DCA techniques for 54 Gy patient group, and 9%, 23% and 34% for 50 Gy patient group, suggesting improvement in delivery efficiency with increasing DCA usage. No significant variations of plan complexity were observed between Hybrid‐DCA and Hybrid‐VMAT (P = 0.46 from Mann–Whitney U‐test), but significant differences were found among DCA, Hybrid and VMAT (P < 0.05). Better agreements between the planned and delivered doses were found with more DCA contributions. By adding DCA components to VMAT planning, hybrid technique offers comparable dosimetry to full VMAT, while increasing delivery efficiency and minimizing MLC complexity.

tion to a DCA plan to increase the dose shaping around the target but maintain many of the advantages of 3D conformal beams. Brain-Lab (BrainLab AG, Feldkirchen, Germany) has developed a HybridArc strategy, which blends aperture-enhanced optimized arcs with several static IMRT-elements at specific intervals. By weighting the contribution of arcs vs IMRT, HybridArc is able to achieve an optimal dose distribution. 9,10 Instead of using IMRT beams, in this study we developed a hybrid planning strategy to incorporate DCA component into VMAT planning. The results of this technique are compared with DCA and VMAT plans in terms of plan quality, plan complexity, treatment delivery efficiency, and the agreement between the planned and delivered doses.

| MATERIALS AND METHODS
Ten NSCLC patients (5 received 54 Gy in 3 fractions and 5 received 50 Gy in 5 fractions) treated with SBRT were retrospectively reviewed in this study. All patients were positioned supine and immobilized with a customized vacuum bag restriction system (Bodyfix, Medical Intelligence Inc) for simulation and subsequent treatments. Motion management of the tumor was achieved with a paddle-based abdominal compression device. All patients underwent a free breathing and ten-phase four-dimensional computer tomography (4DCT) scan on a Philips Brilliance Big Bore CT scanner (Philips, Cleveland, OH). The Philips bellows system was placed around the abdomen to monitor the patient respiratory motion. The free breathing CTs were used for treatment planning and were acquired with fields of view large enough to cover the patient and immobilization devices with 2 mm slice thickness. An internal target volume (ITV) was generated based on the maximum intensity projection (MIP) of 4DCT image, and the planning target volume (PTV) was created by a 5 mm uniform expansion of the ITV. All patients were treated on a Varian TrueBeam STx platform with cone beam CT (CBCT) image guidance.
Four planning techniques were simulated in this study: DCA, Hybrid-DCA (2/3 of the doses from DCA beams), Hybrid-VMAT (2/ 3 of the doses from VMAT beams) and VMAT. The same beam configurations (gantry, collimator and table angles) were used for all four techniques. For the hybrid strategies, DCA beams were conformed to the PTV on the beams-eye-view to deliver a fraction (1/ The maximum point dose and dose-volume constraints of several critical structures are listed in Table 1 for both 54 Gy and 50 Gy protocols. [11][12][13] Selected dose-volume parameters were compared, including the conformity index (CI 50 , ratio of the volume receiving 50% of prescription dose to the PTV volume), V 20 and V 5 (lung volumes receiving 20 Gy and higher, and 5 Gy and higher, respectively) for combined lungs, V 30 Gy (volume receiving 30 Gy and higher) for the chest wall, and D 0.035 cc (dose to 0.035 cc of the volume, a representative of maximum dose) for all other critical structures such as the spinal cord, aorta, trachea, etc.
In this study, the plan complexity was assessed by the MCS,

| RESULTS
The average PTV volume were 36.4 ± 12.3 cc for the 54 Gy patient group and 38.8 ± 15.4 cc for the 50 Gy patient group, respectively. Table 2 shows the tumor locations and target coverages included in this analysis. Table 3     In this study, the overall plan complexity was evaluated by a single metric, MCS (between 0 and 1), for all four planning strategies.
MCS incorporates the leaf sequence variability and aperture area variability into the calculation. As seen in Table 3 and Fig. 1  | 121 treatment if there was a critical structure dose concern, which is a reason of increased MLC modulation in the plan.

CONF LICT OF I NTEREST
The authors declare no conflict of interest.

DEFINITION OF MODULATION METRICS (MCS)
The MSC for VMAT plans was originally defined by McNiven for step-and-shoot IMRT plans, 14 and then modified by Masi in order to apply for VMAT plans. 15 The LSV is defined based on the difference in position for adjacent MLC leaves in the same bank for each control point (CP). The positional variations are considered relative to the maximum MLC variation per bank for each CP. The maximum variation per CP is defined as: Pos max ðCPÞ ¼ hmaxðpos n ∈ N Þ À min ðpos n ∈ N Þi leaf bank ; where N is the number of moving leaves and pos n is the position of leaf n.
The AAV is based on the area defined by opposing MLC leaves in a CP normalized to the maximum area in the arc, which is defined by the maximum apertures for all leaf pairs over all CPs in the arc AAV CP ¼ ∑ A a ¼ 1 hpos a i left bank À hpos a i right bank ∑ A a ¼ 1 hmaxðpos a Þi left bank ∈ arc À hmaxðpos a Þi right bank ∈ arc ; where A is the number of moving leaves in the arc. The MLCs are continuously moving between CP while MU are being delivered, so MCS considers the mean values of LSV CP and AAV CP between adjacent CPs. This product is weighted by the percentage of total MU delivered between the adjacent CPs: where I is the number of CP and MU CPi,i+1 are the MU delivered between two consecutive CPs.
The AA term is defined using the notation from above as AACP ¼ ∑ A a¼1 hpos a i left bank À hpos a i right bank Â w a where w a is the width of leaf a MUarc ! I À 1 :