Commissioning of a Versa HDTM linear accelerator for three commercial treatment planning systems

Abstract In a mixed‐vendor radiation oncology environment, it is advantageous if the department's treatment planning system (TPS) supports the linear accelerators of different vendors. In this publication beam data collection and modeling for the Versa HD linear accelerator in Monaco, Pinnacle, and Eclipse are discussed. In each TPS static field, Intensity‐Modulated Radiation Therapy (IMRT) step and shoot, and Volumetric‐Modulated Arc Therapy (VMAT) plans for flattened and flattening‐filter free photon beams of all available energies were evaluated for field sizes >3 × 3. To compare passing rates, identical beam model validation plans were calculated in each TPS. Eclipse, Monaco, and Pinnacle beam models passed validation measurements in homogeneous materials for a variety of treatment fields, including static, IMRT, and VMAT. In the case of Eclipse, the “dosimetric leaf gap” parameter was found to be critical for passing rates of VMAT plans. The source size parameter plays an important role as well for small fields. In the case of Pinnacle the multileaf collimator offset table needed to be optimized for better VMAT QA results. Each of the investigated treatment planning systems met the criteria to be used clinically in conjunction with Elekta Versa HD linear accelerators. It can be of great advantage to have the option to operate a TPS and linear accelerator from different vendors, as decisions surrounding linear accelerator or TPS purchases are very complicated and not just limited to technical considerations.

The Monaco (Elekta, Atlanta, GA) and Pinnacle (Philips, Fitchburg, WI) TPSs supported Versa HD linear accelerators (Elekta, Atlanata, GA) equipped with the Agility 160MLC and FFF beams soon after market clearance. Starting with TPS version 13.6 MR0.7, Eclipse (Varian, Palo Alto, CA) included the support of planning for the Elekta machine VersaHD with FFF beams as well. The only limitation being that while Eclipse and Pinnacle only support the IMRT step and shoot technique for Versa HD linear accelerators, Monaco also supports the dynamic MLC delivery of IMRT plans. This is not necessarily a major limitation, as it has been shown that dosimetrically the IMRT step and shoot technique and dynamic MLC delivery produce comparable results. The difference is that compared to dynamic MLC delivery the step and shoot technique can be between 15 and 50% slower, while the number of monitor units required is about less. 1,2 It is important to stress the comprehensive program required when acquiring both beam data for model generation as well as recording verification measurements. 3 A model can only be tuned to the quality of its inputs. Similarly, any error in the calibration of the verification system will corrupt the tuning and thus quality of the beam model, regardless of the quality of the original beam data.
In this publication the beam data collection and modeling process for a Versa HD linear accelerator in Monaco, Pinnacle, and Eclipse are discussed. In all three TPSs static field plans, IMRT step and shoot plans, and VMAT plans for flattened and FFF photon beams were evaluated. To compare passing rates, identical beam model validation plans were calculated in all three TPSs.
Beam modeling guides for Monaco, Pinnacle, and Eclipse were carefully reviewed. In an attempt to minimize the amount of data to be collected, data were measured such that it met the beam data requirements of several TPS.

2.A | Equipment
To ensure highest quality of our linear accelerator commissioning data, we followed guidelines and recommendations as published in task group report 106 (TG-106). 4 Profiles and percent depth dose curves (PDDs) were measured with the microDiamond detector type 60019 and large field profiles (20 × 20 cm 2 and larger) with the LA48 linear array (both PTW-Freiburg). The microDiamond has a high spatial resolution, which is critical for correct penumbra modeling of photon and electron beams. 5  Output factors (TSCF) were measured at SSD 90 and 10 cm depth (reference point depth) in water for the same field sizes as profiles and PDDs. Pinnacle did not require any additional collimator factor measurements. For Monaco it was necessary to take additional collimator factor measurements for all field sizes as listed in Table 1. For Eclipse it was necessary to measure a large number of additional output factors (TSCF) for each energy and for wedged and open fields as listed in Table 2. An important caveat for Eclipse is that while the beam model fitting will disregard output factors for field sizes 2 × 2 and below, these factors will still be used by Eclipse during the calculation of small fields. It is thus important to include these factors.
Without small field output factors, Eclipse will calculate small fields with extrapolated values instead of actual acquired data and will give a warning message. Also worth considering is the effects of the location of the MLC and its dual role as a collimator in the X-direction. The Eclipse multisource model is suited to model this. 6 In addition, each TPS required description of calibration conditions (e.g., for SSD 100 cm @ 10 cm and a 10 × 10 field 100 MU = 67.5 cGy for a 6 MV beam). However, it is important that reference T A B L E 1 Required collimator factor measurements for Monaco (for wedged fields collimator factors are only needed for square fields).  T A B L E 2 Additional output factor (TSCF) data (per energy, wedged, and open fields) as required by Eclipse.  Table 4). It is important to note that the DLG value has a huge impact on the passing rates of VMAT and sliding-window    Fig. 1(c)], which had slightly lower pass rates in all planning systems. Results are summarized in Table 5. As can be seen, however, all results would still be clinically acceptable.

Eclipse verification plan results from calculations in Eclipse and
Monaco were found to be clinically acceptable (Table 6).

3.D | Verification of Eclipse VMAT plans
No problems were found with optimizing Eclipse VMAT plans with the Elekta beam model. The optimization process was run through once for each plan to (a) reduce the time required to run multiple optimizations, (b) test the ability of the optimization system to produce clinically acceptable plans without having to modify the initial constraints, and (c) test the ability of the system to reproduce clinically acceptable plans for various energies without further modification.
All Eclipse 6 MV VMAT plans were calculated on the Delta4 phantom and measured with a clinically acceptable passing rate (see Table 7) using a gamma criteria no greater than 3%/3 mm with an acceptable threshold of 90%. All Eclipse 10 MV VMAT plans were delivered onto the Octavius phantom and evaluated using a gamma criteria of 2%/2 mm or 3%/3 mm with an acceptable threshold of 90%. (see Table 7). All results were found to be clinically acceptable as well.

ACKNOWLEDGMENTS
The authors thank Sylvia Spiessen from Varian for her support during this project.