We evaluated the VMAT plans based on SmartArc using a Varian Trilogy linear accelerator; this accelerator is now used clinically for the treatment of endometrial cancer in our department, a complex situation often encountered in the clinic. As compared with the IMRT plan, the VMAT-S plan provided a more homogeneous dose distribution in the PTV and better sparing of the OARs and NT in the medium to high dose region; a slightly larger volume of normal tissue received a radiation dose of 20 Gy. No significant difference was found between the VMAT-S and HT plans. The major benefits of VMAT-S plan were manifested in the faster delivery time and lower MU relative to the IMRT and HT plans. Luca et al.  compared fixed field IMRT with VMAT for cervical cancer as planned/delivered using an Eclipse/Varian linear accelerator. They also found that RapidArc improved dose homogeneity and sparing of the rectum, bladder and small bowel in the medium to high dose region.
The volumes receiving doses of >30 Gy in the bladder, rectum and pelvis bone were reduced using the VMAT plan relative to the IMRT plan, whereas the volumes receiving doses <20 Gy were increased for the bladder and pelvis bone. This was because in IMRT the dose is delivered using relatively few beam angles as compared with VMAT. The improved sparing of the bladder, rectum and pelvis bone at medium to high doses using VMAT as compared with IMRT is expected to further reduce the acute and late toxicities, especially for patients requiring a local boost and concurrent/sequential chemotherapy. This is also relevant to patients not suitable for the high dose rate boost. As an arc-based approach to the delivery of IMRT, VMAT can deliver a more homogeneous dose to the target volume with a greater degree of freedom of intensity modulation. As expected, greater volumes of bowel, pelvic bones and NT received radiation doses ranging from 5–20 Gy, as compared with IMRT. The increased low dose bath effect in the NT and pelvic bones might be explained by the larger and longer target volumes exposed to more radiation beams in the arc pattern of radiation delivery involved in VMAT. Lian et al.  also found that in postoperative endometrial cancer patients the use of HT increased low dose irradiation of the normal tissue and skeleton in pelvic and para-aortic radiotherapy. A greater volume of pelvic bones exposed to a dose of 2–20 Gy could increase the risk of hematologic suppression [14, 15] and bone fracture . A larger volume of NT received a low dose of 2–20 Gy using VMAT-S relative to IMRT. Some concerns have been raised regarding the risk of secondary cancers in NT irradiated to low dose. Given the better sparing of OARs, and the longer life expectancy of older patients with endometrial cancer treated using VMAT-S, its benefits outweigh its pitfalls. Investigation of this issue was beyond the scope of this study, and has previously been addressed and discussed . It is possible that the low dose volume in the pelvic bones and NT could be decreased in the planning process by introducing the corresponding dose volume constraints in VMAT-S and HT. Because the present study was designed to be a comparative dosimetric evaluation of VMAT-S, IMRT and HT plans, we did not use any constraints regarding the pelvic bones and NT, and used the same dose volume objectives and constraints in all three techniques based on our experience and a pilot study. Of course, it is possible that there may be slight differences in the results caused by the different optimization algorithms used in each of the unique planning systems.
VMAT-S and HT provided very similar and highly conformal plans. HT provided a more homogeneous dose distribution in the PTV105 (16.7% vs. 40.5%; P = 0.00), but no significant difference in terms of the HI (1.06 vs. 1.07; P = 0.25). The integral dose delivered to normal tissue was also equivalent using VMAT-S and HT in our study. Delivery of a statistically significant higher integral dose to normal tissue for has been reported for HT relative to VMAT in previous studies [20, 21]. However, the difference was small (approximately 3%). The clinical relevance is very difficult to assess. A study published by D’Souza and Rosen  suggested that the total energy deposited in a patient is relatively independent of treatment planning parameters (such as beam orientation or relative weighting when many beams are used) for deep-seated targets. In addition, because bladder, rectum and bowel, and pelvis bone overlapped with the PTV, their maximum doses were correlated to the minimum dose delivered to the PTV. In the current study, the V50 for bladder, rectum, bowel and pelvis bone were all equivalent among three techniques.
The major benefits of VMAT-S were manifested in the faster treatment delivery time and lower MU as compared with IMRT and HT. The delivery time for IMRT is significantly higher than that for VMAT due to the multiple field arrangement, time to reposition the gantry and mode up signal of the Clinac for every field. The average treatment delivery time was reduced by more than 6 minutes using the VMAT plans as compared with IMRT and HT plans. This reduction in treatment delivery time is clinically relevant in relation to patient comfort and infra-fraction motion. Faster delivery could improve patient adherence to treatment and reduce intra-fractional motion. In addition, the higher delivery efficiency also allowed for more time to carry out image-guided radiotherapy, further reducing the treatment margin and toxicity. More patients could be treated per day using VMAT due to the short delivery time. In addition, the delivery efficiency of the SmartArc plans is higher in terms of requiring less MUs.