On the use of single‐isocenter VMAT plans for SBRT treatment of synchronous multiple lung lesions: Plan quality, treatment efficiency, and early clinical outcomes

Abstract Cone‐beam computed tomography (CT)‐guided volumetric‐modulated arc therapy (VMAT) plans for stereotactic body radiotherapy (SBRT) treatment of synchronous multiple lung lesions with a flattening filter‐free (FFF) beam is a safe and highly effective treatment option for oligometastases lung cancer patients. Fourteen patients with metastatic non–small‐cell lung cancer (NSCLC) lesions (two to five) received a single‐isocenter VMAT SBRT treatment in our clinic. Four‐dimensional (4D) CT‐based treatment plans were generated using advanced AcurosXB‐based dose calculation algorithm using heterogeneity corrections with a single isocenter placed between/among the lesions. Compared to 10X‐FFF and traditional flattened 6X (6X‐FF) beams, 6X‐FFF beam produced highly conformal radiosurgical dose distribution to each target volume, reduced dose to adjacent organs at risk (OAR), and significantly reduced the lung SBRT fraction duration to < 3.5 min/fraction for 54/50 Gy treatments in 3/5 fractions — significantly improving patient convenience and clinic workflow. Early follow‐up CT imaging (mean, 9 months) results show high local control rates (100%) with no acute lung or rib toxicity. Longer clinical follow up in a larger patient cohort is ongoing to further validate the outcomes of this treatment approach.


| INTRODUCTION
With recent advances in radiotherapy technology, stereotactic body radiotherapy (SBRT) treatment has become a standard curative intent treatment for medically inoperable non-small-cell lung cancer (NSCLC) patients. [1][2][3][4][5][6][7] However, some patients with multiple primary or oligometastastic (<5) lung lesions with comorbid illnesses are unable to maintain treatment position adequately for the duration of a SBRT treatment in which individual isocenters are used for each lesion. Furthermore, reducing treatment time would be advantageous in diminishing intrafraction motion errors that accompany longer treatments. 8,9 Compared to a traditional flattened beam with a flat-synchronously using a single-isocenter VMAT-SBRT plan is a fast and efficient treatment technique that is gaining popularity. [16][17][18] As part of our SBRT commissioning for treating multiple lung lesions synchronously via a single-isocenter VMAT plan, we have investigated plan quality and treatment efficiency for different beams and reported early clinical outcomes. The dosimetric differences of traditional flattened 6X-beam (6X-FF) vs FFF beams for a single-lesion lung SBRT treatment have been studied previously, along with the feasibility of treating multiple lung lesions concurrently using a single-isocenter approach. [19][20][21][22][23][24][25][26] However, plan quality evaluation of FFF beams in the synchronous treatment of multiple lung lesions using a single-isocenter VMAT-SBRT plan and the phenomenon of MLC modulation have not yet been investigated. For instance, in a single-isocenter multifocal VMAT lung SBRT setting, adequate tumor coverage of each lesion requires the MLC leaves to travel a longer distance between the lesions. Moreover, due to considerably softer energy spectra of 6X-FFF beam (1.28 MeV) compared to conventional 6X-FF (1.75 MeV), the range of secondary electrons generated by the 6X-FFF beam in the lung will be shorter and potentially provide quicker dose build up at the lung tissue/tumor interface. 27 Previous studies of FFF beams focused primarily on improved treatment efficiency when treating a single lung lesion with an isocenter placed at the tumor center. Those patients who developed multiple lung lesions may not tolerate long treatment times associated with multilesion lung SBRT using individual isocenters for each lesion. Additionally, the change in plan quality attributed to the characteristics of FFF beams while treating multiple lung lesions synchronously using a single-isocenter VMAT plan remains an interesting topic in need of further research, specifically the effect of SBRT on the higher radiosensitivity of nontarget OAR dose. 28 Therefore, this work was undertaken to quantify the impact of 6X-FFF beam implementation in the context of single-isocenter/multilesion VMAT lung SBRT treatment in our clinic and report early clinical findings.

2.A | Patient characteristics and volume delineation
This institutional review board approved retrospective study that includes 14 SBRT patients with 2-5 synchronous metastatic nonsmall-cell lung lesions. The patients were immobilized using the Body Pro-Lok TM platform (CIVCO system, Orange City, IA) in the supine position, arms above their head with abdominal compression. All planning computed tomography (CT) images were acquired on a GE Lightspeed 16 slice CT scanner (General Electric Medical Systems, Waukesha, WI). CT images were acquired with 512 × 512 pixels at 2.5 mm slice thickness. All patients underwent a free breathing scan followed by all 10 phases of a 4D-CT scan using Varian's RPM System (version 1.7). Internal target volumes (ITVs) were delineated on the 3D CT images, referenced to the maximum intensity projection (MIP) images, and the planning target volumes (PTVs) were created by adding a 5 mm uniform margin around the corresponding ITVs.
The critical structures, such as bilateral lungs excluding the ITV (normal lung), spinal cord, ribs, heart, trachea and bronchus, esophagus, and skin, were delineated on the free-breathing CT images in the Eclipse treatment planning system (TPS). The main tumor characteristics of the patients are shown in Table 1. 2.B | Clinical 6X-FFF plans and treatment delivery A single isocenter was placed approximately equidistant to the separate tumors in each patient. Average isocenter to tumor distance was 5.6 ± 1.9 cm (ranged 3.4-9.5 cm). Highly conformal, clinically optimal VMAT treatment plans were generated on the free-breathing 3D-CT scan using 2-6 co/non-coplanar full/partial arcs (5-10°, couch kicks were used for non-coplanar partial arcs) for the Truebeam lin-
Identical beam geometry, planning objectives, and convergence mode were used in the 6X-FFF and 10X-FFF plans including the NTO parameters and ring structures as described above. The 10X-FFF plans received the same target coverage as the clinical 6X-FFF plans.

2.F | Clinical follow-up
Clinical outcomes evaluated include tumor local-control rates, radiation pneumonitis, and rib toxicity. Kaplan-Meier (KM) estimates of time-to-local failure were investigated (XLSTAT, Microsoft Excel).
Patient follow up included physical exam followed by CT scan every 3 months for the first year and then as clinically indicated.

3.A | Treatment plan and delivery parameters
All plans were acceptable per RTOG criteria; however, statistically significant differences were observed in dose distributions and corresponding dose volume histograms when comparing the three VMAT plans (6X-FF, 6X-FFF, and 10X-FFF) for the same patient, as shown in Fig. 1. The corresponding dose volume histogram for the same patient is shown in Fig. 2 Fig. 3. Because of reduced beam modulation and consistently achieving the maximum dose rate of 1400 MU/ min, the average beam-on time for 6X-FFF plan was very similar to that of 10X-FFF plans (see Fig. 3).   Table 3). This could be partly due to the quick dose buildup at tumor-lung interface with 6X-FFF beam (characteristic of 6X-FFF beam), there were less total number of MU required (for same prescribed dose) compared to 6X-FF and 10X-T A B L E 2 Evaluation of plan quality for 6X-FF, 6X-FFF, and 10X-FFF plans for all lung stereotactic body radiotherapy (SBRT) patients treated with single-isocenter/multiple-lesions VMAT lung SBRT. Prescription was 54 Gy (n = 7)/50 Gy (n = 7) in 3/5 fractions. Mean ± SD (P-value) was reported. Modulation factor (MF) = total MU/prescription (cGy). n.s. = not significant. Significant values were highlighted in bold. A potential concern for SBRT treatment of lung lesions is dose spill in the chest wall and ribs, 33 Another concern is the interplay effect of a change in breathing patterns with MLC modulation, gantry rotation, and dose-rate changes during dose delivery that is not likely to average out during relatively short beam-on times. It has been reported that the interplay effect causes insignificant dose blurring when using two or more arcs, in agreement with this study. 40 The change in respiratory patterns between CT simulation and time of treatment has been studied previously and only small changes (within ±3 mm) due to intrafractional and interfractional motion. 41 potentially decreasing the variation of intrafraction motion error and improving the treatment accuracy and efficiency. Compared to traditional 6X-FF plans, overall reduction in beam-on time of 65% and 60% was achieved when using 10X-FFF and 6X-FFF plans, respectively. Additionally, a 12% (0.4 min) reduction in the average beam-on time was found with 10X-FFF plans compared to 6X-FFF plans, but at the cost of a less conformal plan and higher dose to OAR, including rib. Furthermore, due to the larger number of total MU delivered in the SBRT treatment of lung lesions, photonuclear production could be a concern with 10X-FFF beam, although neutron production has been shown to be minimal provided that the treatment plans for each beam required approximately the same number of total MU, yet it was found that higher photoneutron yield per source electron still existent with 10X-FFF beam. 43 Because of this dosimetric study, 6X-FFF plan was the choice in our clinic for lung SBRT treatment with smaller number of total MU, higher plan quality, and lower dose to OAR, including the treatment of synchronous multifocal lung lesions (see Table 2). Moreover, our early clinical follow-up results (mean, 9 months) are encouraging with 100% local tumor-control rate and no reported treatment-related acute toxicity.

| CONCLUSION
6X-FFF beam for SBRT treatment of multiple lung lesions using a single-isocenter VMAT plan yielded superior plan quality and faster treatment delivery compared to conventional 6X-FF beamperhaps improving clinic efficiency and patient comfort. It has been observed that utilizing 10X-FFF beam could marginally improve the delivery efficiency and skin dose (only for selected patients) but at the cost of inferior dose conformity, gradient indices (higher intermediate dose spill), and higher dose to other OAR. Thus, 6X-FFF beam was chosen in our clinic for lung SBRT patients including synchronous treatment of multiple lung lesions via a single-isocenter VMAT plan. The single-isocenter VMAT technique was fast, effective, and well tolerated by all patients, improving patient compliance and potentially reducing the amount of intrafraction motion errors for well-suited patients. Our early clinical outcomes are promising in terms of local control rates with no acute side effects to lung or ribs and potentially improving the patient's quality of life. Longer clinical follow up of these patients is underway.

CONF LICT OF I NTEREST
None.